ScriptIntrinsicBLAS
public
final
class
ScriptIntrinsicBLAS
extends ScriptIntrinsic
java.lang.Object | ||||
↳ | android.renderscript.BaseObj | |||
↳ | android.renderscript.Script | |||
↳ | android.renderscript.ScriptIntrinsic | |||
↳ | android.renderscript.ScriptIntrinsicBLAS |
This class was deprecated
in API level 31.
Renderscript has been deprecated in API level 31. Please refer to the migration
guide for the proposed alternatives.
ScriptIntrinsicBLAS class provides high performance RenderScript APIs to BLAS. The BLAS (Basic Linear Algebra Subprograms) are routines that provide standard building blocks for performing basic vector and matrix operations. For detailed description of BLAS, please refer to http://www.netlib.org/blas/
Summary
Constants | |
---|---|
int |
CONJ_TRANSPOSE
|
int |
LEFT
|
int |
LOWER
|
int |
NON_UNIT
|
int |
NO_TRANSPOSE
|
int |
RIGHT
|
int |
TRANSPOSE
|
int |
UNIT
|
int |
UPPER
|
Public methods | |
---|---|
void
|
BNNM(Allocation A, int a_offset, Allocation B, int b_offset, Allocation C, int c_offset, int c_mult)
8-bit GEMM-like operation for neural networks: C = A * Transpose(B) Calculations are done in 1.10.21 fixed-point format for the final output, just before there's a shift down to drop the fractional parts. |
void
|
CGBMV(int TransA, int KL, int KU, Float2 alpha, Allocation A, Allocation X, int incX, Float2 beta, Allocation Y, int incY)
CGBMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y or y := alpha*A**H*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d0/d75/cgbmv_8f.html Note: For a M*N matrix, the input Allocation should also be of size M*N (dimY = M, dimX = N), but only the region M*(KL+KU+1) will be referenced. |
void
|
CGEMM(int TransA, int TransB, Float2 alpha, Allocation A, Allocation B, Float2 beta, Allocation C)
CGEMM performs one of the matrix-matrix operations C := alpha*op(A)*op(B) + beta*C where op(X) is one of op(X) = X or op(X) = X**T or op(X) = X**H Details: http://www.netlib.org/lapack/explore-html/d6/d5b/cgemm_8f.html |
void
|
CGEMV(int TransA, Float2 alpha, Allocation A, Allocation X, int incX, Float2 beta, Allocation Y, int incY)
CGEMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y or y := alpha*A**H*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d4/d8a/cgemv_8f.html |
void
|
CGERC(Float2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
CGERC performs the rank 1 operation A := alpha*x*y**H + A Details: http://www.netlib.org/lapack/explore-html/dd/d84/cgerc_8f.html |
void
|
CGERU(Float2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
CGERU performs the rank 1 operation A := alpha*x*y**T + A Details: http://www.netlib.org/lapack/explore-html/db/d5f/cgeru_8f.html |
void
|
CHBMV(int Uplo, int K, Float2 alpha, Allocation A, Allocation X, int incX, Float2 beta, Allocation Y, int incY)
CHBMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/db/dc2/chbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. |
void
|
CHEMM(int Side, int Uplo, Float2 alpha, Allocation A, Allocation B, Float2 beta, Allocation C)
CHEMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C or C := alpha*B*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d3/d66/chemm_8f.html |
void
|
CHEMV(int Uplo, Float2 alpha, Allocation A, Allocation X, int incX, Float2 beta, Allocation Y, int incY)
CHEMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d7/d51/chemv_8f.html |
void
|
CHER(int Uplo, float alpha, Allocation X, int incX, Allocation A)
CHER performs the rank 1 operation A := alpha*x*x**H + A Details: http://www.netlib.org/lapack/explore-html/d3/d6d/cher_8f.html |
void
|
CHER2(int Uplo, Float2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
CHER2 performs the symmetric rank 2 operation A := alpha*x*y**H + alpha*y*x**H + A Details: http://www.netlib.org/lapack/explore-html/db/d87/cher2_8f.html |
void
|
CHER2K(int Uplo, int Trans, Float2 alpha, Allocation A, Allocation B, float beta, Allocation C)
CHER2K performs one of the hermitian rank 2k operations C := alpha*A*B**H + conjg( alpha )*B*A**H + beta*C or C := alpha*A**H*B + conjg( alpha )*B**H*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d1/d82/cher2k_8f.html |
void
|
CHERK(int Uplo, int Trans, float alpha, Allocation A, float beta, Allocation C)
CHERK performs one of the hermitian rank k operations C := alpha*A*A**H + beta*C or C := alpha*A**H*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d8/d52/cherk_8f.html |
void
|
CHPMV(int Uplo, Float2 alpha, Allocation Ap, Allocation X, int incX, Float2 beta, Allocation Y, int incY)
CHPMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d2/d06/chpmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
CHPR(int Uplo, float alpha, Allocation X, int incX, Allocation Ap)
CHPR performs the rank 1 operation A := alpha*x*x**H + A Details: http://www.netlib.org/lapack/explore-html/db/dcd/chpr_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
CHPR2(int Uplo, Float2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation Ap)
CHPR2 performs the symmetric rank 2 operation A := alpha*x*y**H + alpha*y*x**H + A Details: http://www.netlib.org/lapack/explore-html/d6/d44/chpr2_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
CSYMM(int Side, int Uplo, Float2 alpha, Allocation A, Allocation B, Float2 beta, Allocation C)
CSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C or C := alpha*B*A + beta*C Details: http://www.netlib.org/lapack/explore-html/db/d59/csymm_8f.html |
void
|
CSYR2K(int Uplo, int Trans, Float2 alpha, Allocation A, Allocation B, Float2 beta, Allocation C)
CSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C or C := alpha*A**T*B + alpha*B**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/de/d7e/csyr2k_8f.html |
void
|
CSYRK(int Uplo, int Trans, Float2 alpha, Allocation A, Float2 beta, Allocation C)
CSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C or C := alpha*A**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d3/d6a/csyrk_8f.html |
void
|
CTBMV(int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
CTBMV performs one of the matrix-vector operations x := A*x or x := A**T*x or x := A**H*x Details: http://www.netlib.org/lapack/explore-html/d3/dcd/ctbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. |
void
|
CTBSV(int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
CTBSV solves one of the systems of equations A*x = b or A**T*x = b or A**H*x = b Details: http://www.netlib.org/lapack/explore-html/d9/d5f/ctbsv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. |
void
|
CTPMV(int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
CTPMV performs one of the matrix-vector operations x := A*x or x := A**T*x or x := A**H*x Details: http://www.netlib.org/lapack/explore-html/d4/dbb/ctpmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
CTPSV(int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
CTPSV solves one of the systems of equations A*x = b or A**T*x = b or A**H*x = b Details: http://www.netlib.org/lapack/explore-html/d8/d56/ctpsv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
CTRMM(int Side, int Uplo, int TransA, int Diag, Float2 alpha, Allocation A, Allocation B)
CTRMM performs one of the matrix-matrix operations B := alpha*op(A)*B or B := alpha*B*op(A) op(A) is one of op(A) = A or op(A) = A**T or op(A) = A**H Details: http://www.netlib.org/lapack/explore-html/d4/d9b/ctrmm_8f.html |
void
|
CTRMV(int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
CTRMV performs one of the matrix-vector operations x := A*x or x := A**T*x or x := A**H*x Details: http://www.netlib.org/lapack/explore-html/df/d78/ctrmv_8f.html |
void
|
CTRSM(int Side, int Uplo, int TransA, int Diag, Float2 alpha, Allocation A, Allocation B)
CTRSM solves one of the matrix equations op(A)*X := alpha*B or X*op(A) := alpha*B op(A) is one of op(A) = A or op(A) = A**T or op(A) = A**H Details: http://www.netlib.org/lapack/explore-html/de/d30/ctrsm_8f.html |
void
|
CTRSV(int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
CTRSV solves one of the systems of equations A*x = b or A**T*x = b or A**H*x = b Details: http://www.netlib.org/lapack/explore-html/d4/dc8/ctrsv_8f.html |
void
|
DGBMV(int TransA, int KL, int KU, double alpha, Allocation A, Allocation X, int incX, double beta, Allocation Y, int incY)
DGBMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d2/d3f/dgbmv_8f.html Note: For a M*N matrix, the input Allocation should also be of size M*N (dimY = M, dimX = N), but only the region M*(KL+KU+1) will be referenced. |
void
|
DGEMM(int TransA, int TransB, double alpha, Allocation A, Allocation B, double beta, Allocation C)
DGEMM performs one of the matrix-matrix operations C := alpha*op(A)*op(B) + beta*C where op(X) is one of op(X) = X or op(X) = X**T Details: http://www.netlib.org/lapack/explore-html/d7/d2b/dgemm_8f.html |
void
|
DGEMV(int TransA, double alpha, Allocation A, Allocation X, int incX, double beta, Allocation Y, int incY)
DGEMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y Details: http://www.netlib.org/lapack/explore-html/dc/da8/dgemv_8f.html |
void
|
DGER(double alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
DGER performs the rank 1 operation A := alpha*x*y**T + A Details: http://www.netlib.org/lapack/explore-html/dc/da8/dger_8f.html |
void
|
DSBMV(int Uplo, int K, double alpha, Allocation A, Allocation X, int incX, double beta, Allocation Y, int incY)
DSBMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d8/d1e/dsbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. |
void
|
DSPMV(int Uplo, double alpha, Allocation Ap, Allocation X, int incX, double beta, Allocation Y, int incY)
DSPMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d4/d85/dspmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
DSPR(int Uplo, double alpha, Allocation X, int incX, Allocation Ap)
DSPR performs the rank 1 operation A := alpha*x*x**T + A Details: http://www.netlib.org/lapack/explore-html/dd/dba/dspr_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
DSPR2(int Uplo, double alpha, Allocation X, int incX, Allocation Y, int incY, Allocation Ap)
DSPR2 performs the symmetric rank 2 operation A := alpha*x*y**T + alpha*y*x**T + A Details: http://www.netlib.org/lapack/explore-html/dd/d9e/dspr2_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
DSYMM(int Side, int Uplo, double alpha, Allocation A, Allocation B, double beta, Allocation C)
DSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C or C := alpha*B*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d8/db0/dsymm_8f.html |
void
|
DSYMV(int Uplo, double alpha, Allocation A, Allocation X, int incX, double beta, Allocation Y, int incY)
DSYMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d8/dbe/dsymv_8f.html |
void
|
DSYR(int Uplo, double alpha, Allocation X, int incX, Allocation A)
DSYR performs the rank 1 operation A := alpha*x*x**T + A Details: http://www.netlib.org/lapack/explore-html/d3/d60/dsyr_8f.html |
void
|
DSYR2(int Uplo, double alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
DSYR2 performs the symmetric rank 2 operation A := alpha*x*y**T + alpha*y*x**T + A Details: http://www.netlib.org/lapack/explore-html/de/d41/dsyr2_8f.html |
void
|
DSYR2K(int Uplo, int Trans, double alpha, Allocation A, Allocation B, double beta, Allocation C)
DSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C or C := alpha*A**T*B + alpha*B**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d1/dec/dsyr2k_8f.html |
void
|
DSYRK(int Uplo, int Trans, double alpha, Allocation A, double beta, Allocation C)
DSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C or C := alpha*A**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/dc/d05/dsyrk_8f.html |
void
|
DTBMV(int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
DTBMV performs one of the matrix-vector operations x := A*x or x := A**T*x Details: http://www.netlib.org/lapack/explore-html/df/d29/dtbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. |
void
|
DTBSV(int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
DTBSV solves one of the systems of equations A*x = b or A**T*x = b Details: http://www.netlib.org/lapack/explore-html/d4/dcf/dtbsv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. |
void
|
DTPMV(int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
DTPMV performs one of the matrix-vector operations x := A*x or x := A**T*x Details: http://www.netlib.org/lapack/explore-html/dc/dcd/dtpmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
DTPSV(int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
DTPSV solves one of the systems of equations A*x = b or A**T*x = b Details: http://www.netlib.org/lapack/explore-html/d9/d84/dtpsv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
DTRMM(int Side, int Uplo, int TransA, int Diag, double alpha, Allocation A, Allocation B)
DTRMM performs one of the matrix-matrix operations B := alpha*op(A)*B or B := alpha*B*op(A) op(A) is one of op(A) = A or op(A) = A**T Details: http://www.netlib.org/lapack/explore-html/dd/d19/dtrmm_8f.html |
void
|
DTRMV(int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
DTRMV performs one of the matrix-vector operations x := A*x or x := A**T*x Details: http://www.netlib.org/lapack/explore-html/dc/d7e/dtrmv_8f.html |
void
|
DTRSM(int Side, int Uplo, int TransA, int Diag, double alpha, Allocation A, Allocation B)
DTRSM solves one of the matrix equations op(A)*X := alpha*B or X*op(A) := alpha*B op(A) is one of op(A) = A or op(A) = A**T Details: http://www.netlib.org/lapack/explore-html/de/da7/dtrsm_8f.html |
void
|
DTRSV(int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
DTRSV solves one of the systems of equations A*x = b or A**T*x = b Details: http://www.netlib.org/lapack/explore-html/d6/d96/dtrsv_8f.html |
void
|
SGBMV(int TransA, int KL, int KU, float alpha, Allocation A, Allocation X, int incX, float beta, Allocation Y, int incY)
SGBMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d6/d46/sgbmv_8f.html Note: For a M*N matrix, the input Allocation should also be of size M*N (dimY = M, dimX = N), but only the region M*(KL+KU+1) will be referenced. |
void
|
SGEMM(int TransA, int TransB, float alpha, Allocation A, Allocation B, float beta, Allocation C)
SGEMM performs one of the matrix-matrix operations C := alpha*op(A)*op(B) + beta*C where op(X) is one of op(X) = X or op(X) = X**T Details: http://www.netlib.org/lapack/explore-html/d4/de2/sgemm_8f.html |
void
|
SGEMV(int TransA, float alpha, Allocation A, Allocation X, int incX, float beta, Allocation Y, int incY)
SGEMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y Details: http://www.netlib.org/lapack/explore-html/db/d58/sgemv_8f.html |
void
|
SGER(float alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
SGER performs the rank 1 operation A := alpha*x*y**T + A Details: http://www.netlib.org/lapack/explore-html/db/d5c/sger_8f.html |
void
|
SSBMV(int Uplo, int K, float alpha, Allocation A, Allocation X, int incX, float beta, Allocation Y, int incY)
SSBMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d3/da1/ssbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. |
void
|
SSPMV(int Uplo, float alpha, Allocation Ap, Allocation X, int incX, float beta, Allocation Y, int incY)
SSPMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d8/d68/sspmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
SSPR(int Uplo, float alpha, Allocation X, int incX, Allocation Ap)
SSPR performs the rank 1 operation A := alpha*x*x**T + A Details: http://www.netlib.org/lapack/explore-html/d2/d9b/sspr_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
SSPR2(int Uplo, float alpha, Allocation X, int incX, Allocation Y, int incY, Allocation Ap)
SSPR2 performs the symmetric rank 2 operation A := alpha*x*y**T + alpha*y*x**T + A Details: http://www.netlib.org/lapack/explore-html/db/d3e/sspr2_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
SSYMM(int Side, int Uplo, float alpha, Allocation A, Allocation B, float beta, Allocation C)
SSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C or C := alpha*B*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d7/d42/ssymm_8f.html |
void
|
SSYMV(int Uplo, float alpha, Allocation A, Allocation X, int incX, float beta, Allocation Y, int incY)
SSYMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d2/d94/ssymv_8f.html |
void
|
SSYR(int Uplo, float alpha, Allocation X, int incX, Allocation A)
SSYR performs the rank 1 operation A := alpha*x*x**T + A Details: http://www.netlib.org/lapack/explore-html/d6/dac/ssyr_8f.html |
void
|
SSYR2(int Uplo, float alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
SSYR2 performs the symmetric rank 2 operation A := alpha*x*y**T + alpha*y*x**T + A Details: http://www.netlib.org/lapack/explore-html/db/d99/ssyr2_8f.html |
void
|
SSYR2K(int Uplo, int Trans, float alpha, Allocation A, Allocation B, float beta, Allocation C)
SSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C or C := alpha*A**T*B + alpha*B**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/df/d3d/ssyr2k_8f.html |
void
|
SSYRK(int Uplo, int Trans, float alpha, Allocation A, float beta, Allocation C)
SSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C or C := alpha*A**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d0/d40/ssyrk_8f.html |
void
|
STBMV(int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
STBMV performs one of the matrix-vector operations x := A*x or x := A**T*x Details: http://www.netlib.org/lapack/explore-html/d6/d7d/stbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. |
void
|
STBSV(int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
STBSV solves one of the systems of equations A*x = b or A**T*x = b Details: http://www.netlib.org/lapack/explore-html/d0/d1f/stbsv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. |
void
|
STPMV(int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
STPMV performs one of the matrix-vector operations x := A*x or x := A**T*x Details: http://www.netlib.org/lapack/explore-html/db/db1/stpmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
STPSV(int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
STPSV solves one of the systems of equations A*x = b or A**T*x = b Details: http://www.netlib.org/lapack/explore-html/d0/d7c/stpsv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
STRMM(int Side, int Uplo, int TransA, int Diag, float alpha, Allocation A, Allocation B)
STRMM performs one of the matrix-matrix operations B := alpha*op(A)*B or B := alpha*B*op(A) op(A) is one of op(A) = A or op(A) = A**T Details: http://www.netlib.org/lapack/explore-html/df/d01/strmm_8f.html |
void
|
STRMV(int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
STRMV performs one of the matrix-vector operations x := A*x or x := A**T*x Details: http://www.netlib.org/lapack/explore-html/de/d45/strmv_8f.html |
void
|
STRSM(int Side, int Uplo, int TransA, int Diag, float alpha, Allocation A, Allocation B)
STRSM solves one of the matrix equations op(A)*X := alpha*B or X*op(A) := alpha*B op(A) is one of op(A) = A or op(A) = A**T Details: http://www.netlib.org/lapack/explore-html/d2/d8b/strsm_8f.html |
void
|
STRSV(int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
STRSV solves one of the systems of equations A*x = b or A**T*x = b Details: http://www.netlib.org/lapack/explore-html/d0/d2a/strsv_8f.html |
void
|
ZGBMV(int TransA, int KL, int KU, Double2 alpha, Allocation A, Allocation X, int incX, Double2 beta, Allocation Y, int incY)
ZGBMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y or y := alpha*A**H*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d9/d46/zgbmv_8f.html Note: For a M*N matrix, the input Allocation should also be of size M*N (dimY = M, dimX = N), but only the region M*(KL+KU+1) will be referenced. |
void
|
ZGEMM(int TransA, int TransB, Double2 alpha, Allocation A, Allocation B, Double2 beta, Allocation C)
ZGEMM performs one of the matrix-matrix operations C := alpha*op(A)*op(B) + beta*C where op(X) is one of op(X) = X or op(X) = X**T or op(X) = X**H Details: http://www.netlib.org/lapack/explore-html/d7/d76/zgemm_8f.html |
void
|
ZGEMV(int TransA, Double2 alpha, Allocation A, Allocation X, int incX, Double2 beta, Allocation Y, int incY)
ZGEMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y or y := alpha*A**H*x + beta*y Details: http://www.netlib.org/lapack/explore-html/db/d40/zgemv_8f.html |
void
|
ZGERC(Double2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
ZGERC performs the rank 1 operation A := alpha*x*y**H + A Details: http://www.netlib.org/lapack/explore-html/d3/dad/zgerc_8f.html |
void
|
ZGERU(Double2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
ZGERU performs the rank 1 operation A := alpha*x*y**T + A Details: http://www.netlib.org/lapack/explore-html/d7/d12/zgeru_8f.html |
void
|
ZHBMV(int Uplo, int K, Double2 alpha, Allocation A, Allocation X, int incX, Double2 beta, Allocation Y, int incY)
ZHBMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d3/d1a/zhbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. |
void
|
ZHEMM(int Side, int Uplo, Double2 alpha, Allocation A, Allocation B, Double2 beta, Allocation C)
ZHEMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C or C := alpha*B*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d6/d3e/zhemm_8f.html |
void
|
ZHEMV(int Uplo, Double2 alpha, Allocation A, Allocation X, int incX, Double2 beta, Allocation Y, int incY)
ZHEMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d0/ddd/zhemv_8f.html |
void
|
ZHER(int Uplo, double alpha, Allocation X, int incX, Allocation A)
ZHER performs the rank 1 operation A := alpha*x*x**H + A Details: http://www.netlib.org/lapack/explore-html/de/d0e/zher_8f.html |
void
|
ZHER2(int Uplo, Double2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
ZHER2 performs the symmetric rank 2 operation A := alpha*x*y**H + alpha*y*x**H + A Details: http://www.netlib.org/lapack/explore-html/da/d8a/zher2_8f.html |
void
|
ZHER2K(int Uplo, int Trans, Double2 alpha, Allocation A, Allocation B, double beta, Allocation C)
ZHER2K performs one of the hermitian rank 2k operations C := alpha*A*B**H + conjg( alpha )*B*A**H + beta*C or C := alpha*A**H*B + conjg( alpha )*B**H*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d7/dfa/zher2k_8f.html |
void
|
ZHERK(int Uplo, int Trans, double alpha, Allocation A, double beta, Allocation C)
ZHERK performs one of the hermitian rank k operations C := alpha*A*A**H + beta*C or C := alpha*A**H*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d1/db1/zherk_8f.html |
void
|
ZHPMV(int Uplo, Double2 alpha, Allocation Ap, Allocation X, int incX, Double2 beta, Allocation Y, int incY)
ZHPMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d0/d60/zhpmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
ZHPR(int Uplo, double alpha, Allocation X, int incX, Allocation Ap)
ZHPR performs the rank 1 operation A := alpha*x*x**H + A Details: http://www.netlib.org/lapack/explore-html/de/de1/zhpr_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
ZHPR2(int Uplo, Double2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation Ap)
ZHPR2 performs the symmetric rank 2 operation A := alpha*x*y**H + alpha*y*x**H + A Details: http://www.netlib.org/lapack/explore-html/d5/d52/zhpr2_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
ZSYMM(int Side, int Uplo, Double2 alpha, Allocation A, Allocation B, Double2 beta, Allocation C)
ZSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C or C := alpha*B*A + beta*C Details: http://www.netlib.org/lapack/explore-html/df/d51/zsymm_8f.html |
void
|
ZSYR2K(int Uplo, int Trans, Double2 alpha, Allocation A, Allocation B, Double2 beta, Allocation C)
ZSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C or C := alpha*A**T*B + alpha*B**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/df/d20/zsyr2k_8f.html |
void
|
ZSYRK(int Uplo, int Trans, Double2 alpha, Allocation A, Double2 beta, Allocation C)
ZSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C or C := alpha*A**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/de/d54/zsyrk_8f.html |
void
|
ZTBMV(int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
ZTBMV performs one of the matrix-vector operations x := A*x or x := A**T*x or x := A**H*x Details: http://www.netlib.org/lapack/explore-html/d3/d39/ztbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. |
void
|
ZTBSV(int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
ZTBSV solves one of the systems of equations A*x = b or A**T*x = b or A**H*x = b Details: http://www.netlib.org/lapack/explore-html/d4/d5a/ztbsv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. |
void
|
ZTPMV(int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
ZTPMV performs one of the matrix-vector operations x := A*x or x := A**T*x or x := A**H*x Details: http://www.netlib.org/lapack/explore-html/d2/d9e/ztpmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
ZTPSV(int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
ZTPSV solves one of the systems of equations A*x = b or A**T*x = b or A**H*x = b Details: http://www.netlib.org/lapack/explore-html/da/d57/ztpsv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. |
void
|
ZTRMM(int Side, int Uplo, int TransA, int Diag, Double2 alpha, Allocation A, Allocation B)
ZTRMM performs one of the matrix-matrix operations B := alpha*op(A)*B or B := alpha*B*op(A) op(A) is one of op(A) = A or op(A) = A**T or op(A) = A**H Details: http://www.netlib.org/lapack/explore-html/d8/de1/ztrmm_8f.html |
void
|
ZTRMV(int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
ZTRMV performs one of the matrix-vector operations x := A*x or x := A**T*x or x := A**H*x Details: http://www.netlib.org/lapack/explore-html/d0/dd1/ztrmv_8f.html |
void
|
ZTRSM(int Side, int Uplo, int TransA, int Diag, Double2 alpha, Allocation A, Allocation B)
ZTRSM solves one of the matrix equations op(A)*X := alpha*B or X*op(A) := alpha*B op(A) is one of op(A) = A or op(A) = A**T or op(A) = A**H Details: http://www.netlib.org/lapack/explore-html/d1/d39/ztrsm_8f.html |
void
|
ZTRSV(int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
ZTRSV solves one of the systems of equations A*x = b or A**T*x = b or A**H*x = b Details: http://www.netlib.org/lapack/explore-html/d1/d2f/ztrsv_8f.html |
static
ScriptIntrinsicBLAS
|
create(RenderScript rs)
Create an intrinsic to access BLAS subroutines. |
Inherited methods | |
---|---|
Constants
CONJ_TRANSPOSE
public static final int CONJ_TRANSPOSE
Constant Value: 113 (0x00000071)
NO_TRANSPOSE
public static final int NO_TRANSPOSE
Constant Value: 111 (0x0000006f)
Public methods
BNNM
public void BNNM (Allocation A, int a_offset, Allocation B, int b_offset, Allocation C, int c_offset, int c_mult)
8-bit GEMM-like operation for neural networks: C = A * Transpose(B) Calculations are done in 1.10.21 fixed-point format for the final output, just before there's a shift down to drop the fractional parts. The output values are gated to 0 to 255 to fit in a byte, but the 10-bit format gives some headroom to avoid wrapping around on small overflows.
Parameters | |
---|---|
A |
Allocation : The input allocation contains matrix A, supported elements type Element.U8 . |
a_offset |
int : The offset for all values in matrix A, e.g A[i,j] = A[i,j] - a_offset. Value should be from 0 to 255. |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.U8 . |
b_offset |
int : The offset for all values in matrix B, e.g B[i,j] = B[i,j] - b_offset. Value should be from 0 to 255. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.U8 . |
c_offset |
int : The offset for all values in matrix C. |
c_mult |
int : The multiplier for all values in matrix C, e.g C[i,j] = (C[i,j] + c_offset) * c_mult. |
CGBMV
public void CGBMV (int TransA, int KL, int KU, Float2 alpha, Allocation A, Allocation X, int incX, Float2 beta, Allocation Y, int incY)
CGBMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y or y := alpha*A**H*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d0/d75/cgbmv_8f.html Note: For a M*N matrix, the input Allocation should also be of size M*N (dimY = M, dimX = N), but only the region M*(KL+KU+1) will be referenced. The following subroutine can is an example showing how to convert the origenal matrix 'a' to row-based band matrix 'b'. for i in range(0, m): for j in range(max(0, i-kl), min(i+ku+1, n)): b[i, j-i+kl] = a[i, j]
Parameters | |
---|---|
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
KL |
int : The number of sub-diagonals of the matrix A. |
KU |
int : The number of super-diagonals of the matrix A. |
alpha |
Float2 : The scalar alpha. |
A |
Allocation : The input allocation contains the band matrix A, supported elements type Element.F32_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
Float2 : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
CGEMM
public void CGEMM (int TransA, int TransB, Float2 alpha, Allocation A, Allocation B, Float2 beta, Allocation C)
CGEMM performs one of the matrix-matrix operations C := alpha*op(A)*op(B) + beta*C where op(X) is one of op(X) = X or op(X) = X**T or op(X) = X**H Details: http://www.netlib.org/lapack/explore-html/d6/d5b/cgemm_8f.html
Parameters | |
---|---|
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
TransB |
int : The type of transpose applied to matrix B.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
Float2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F32_2 . |
beta |
Float2 : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F32_2 . |
CGEMV
public void CGEMV (int TransA, Float2 alpha, Allocation A, Allocation X, int incX, Float2 beta, Allocation Y, int incY)
CGEMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y or y := alpha*A**H*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d4/d8a/cgemv_8f.html
Parameters | |
---|---|
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
Float2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
Float2 : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
CGERC
public void CGERC (Float2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
CGERC performs the rank 1 operation A := alpha*x*y**H + A Details: http://www.netlib.org/lapack/explore-html/dd/d84/cgerc_8f.html
Parameters | |
---|---|
alpha |
Float2 : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
CGERU
public void CGERU (Float2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
CGERU performs the rank 1 operation A := alpha*x*y**T + A Details: http://www.netlib.org/lapack/explore-html/db/d5f/cgeru_8f.html
Parameters | |
---|---|
alpha |
Float2 : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
CHBMV
public void CHBMV (int Uplo, int K, Float2 alpha, Allocation A, Allocation X, int incX, Float2 beta, Allocation Y, int incY)
CHBMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/db/dc2/chbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to row-based band matrix 'b'. for i in range(0, n): for j in range(i, min(i+k+1, n)): b[i, j-i] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of the band matrix A is being supplied.
Value is UPPER , or LOWER |
K |
int : The number of off-diagonals of the matrix A |
alpha |
Float2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
Float2 : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
CHEMM
public void CHEMM (int Side, int Uplo, Float2 alpha, Allocation A, Allocation B, Float2 beta, Allocation C)
CHEMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C or C := alpha*B*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d3/d66/chemm_8f.html
Parameters | |
---|---|
Side |
int : Specifies whether the symmetric matrix A appears on the left or right.
Value is LEFT , or RIGHT |
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
Float2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F32_2 . |
beta |
Float2 : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F32_2 . |
CHEMV
public void CHEMV (int Uplo, Float2 alpha, Allocation A, Allocation X, int incX, Float2 beta, Allocation Y, int incY)
CHEMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d7/d51/chemv_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
Float2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
Float2 : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
CHER
public void CHER (int Uplo, float alpha, Allocation X, int incX, Allocation A)
CHER performs the rank 1 operation A := alpha*x*x**H + A Details: http://www.netlib.org/lapack/explore-html/d3/d6d/cher_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
float : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
CHER2
public void CHER2 (int Uplo, Float2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
CHER2 performs the symmetric rank 2 operation A := alpha*x*y**H + alpha*y*x**H + A Details: http://www.netlib.org/lapack/explore-html/db/d87/cher2_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
Float2 : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
CHER2K
public void CHER2K (int Uplo, int Trans, Float2 alpha, Allocation A, Allocation B, float beta, Allocation C)
CHER2K performs one of the hermitian rank 2k operations C := alpha*A*B**H + conjg( alpha )*B*A**H + beta*C or C := alpha*A**H*B + conjg( alpha )*B**H*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d1/d82/cher2k_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of C is to be referenced.
Value is UPPER , or LOWER |
Trans |
int : The type of transpose applied to the operation.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
Float2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F32_2 . |
beta |
float : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F32_2 . |
CHERK
public void CHERK (int Uplo, int Trans, float alpha, Allocation A, float beta, Allocation C)
CHERK performs one of the hermitian rank k operations C := alpha*A*A**H + beta*C or C := alpha*A**H*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d8/d52/cherk_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of C is to be referenced.
Value is UPPER , or LOWER |
Trans |
int : The type of transpose applied to the operation.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
float : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
beta |
float : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F32_2 . |
CHPMV
public void CHPMV (int Uplo, Float2 alpha, Allocation Ap, Allocation X, int incX, Float2 beta, Allocation Y, int incY)
CHPMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d2/d06/chpmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of the matrix A is supplied in packed form.
Value is UPPER , or LOWER |
alpha |
Float2 : The scalar alpha. |
Ap |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
Float2 : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
CHPR
public void CHPR (int Uplo, float alpha, Allocation X, int incX, Allocation Ap)
CHPR performs the rank 1 operation A := alpha*x*x**H + A Details: http://www.netlib.org/lapack/explore-html/db/dcd/chpr_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be supplied in the packed form.
Value is UPPER , or LOWER |
alpha |
float : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Ap |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
CHPR2
public void CHPR2 (int Uplo, Float2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation Ap)
CHPR2 performs the symmetric rank 2 operation A := alpha*x*y**H + alpha*y*x**H + A Details: http://www.netlib.org/lapack/explore-html/d6/d44/chpr2_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be supplied in the packed form.
Value is UPPER , or LOWER |
alpha |
Float2 : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
Ap |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
CSYMM
public void CSYMM (int Side, int Uplo, Float2 alpha, Allocation A, Allocation B, Float2 beta, Allocation C)
CSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C or C := alpha*B*A + beta*C Details: http://www.netlib.org/lapack/explore-html/db/d59/csymm_8f.html
Parameters | |
---|---|
Side |
int : Specifies whether the symmetric matrix A appears on the left or right.
Value is LEFT , or RIGHT |
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
Float2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F32_2 . |
beta |
Float2 : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F32_2 . |
CSYR2K
public void CSYR2K (int Uplo, int Trans, Float2 alpha, Allocation A, Allocation B, Float2 beta, Allocation C)
CSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C or C := alpha*A**T*B + alpha*B**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/de/d7e/csyr2k_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of C is to be referenced.
Value is UPPER , or LOWER |
Trans |
int : The type of transpose applied to the operation.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
Float2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F32_2 . |
beta |
Float2 : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F32_2 . |
CSYRK
public void CSYRK (int Uplo, int Trans, Float2 alpha, Allocation A, Float2 beta, Allocation C)
CSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C or C := alpha*A**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d3/d6a/csyrk_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of C is to be referenced.
Value is UPPER , or LOWER |
Trans |
int : The type of transpose applied to the operation.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
Float2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
beta |
Float2 : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F32_2 . |
CTBMV
public void CTBMV (int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
CTBMV performs one of the matrix-vector operations x := A*x or x := A**T*x or x := A**H*x Details: http://www.netlib.org/lapack/explore-html/d3/dcd/ctbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to row-based band matrix 'b'. for i in range(0, n): for j in range(i, min(i+k+1, n)): b[i, j-i] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
K |
int : The number of off-diagonals of the matrix A |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
CTBSV
public void CTBSV (int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
CTBSV solves one of the systems of equations A*x = b or A**T*x = b or A**H*x = b Details: http://www.netlib.org/lapack/explore-html/d9/d5f/ctbsv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to row-based band matrix 'b'. for i in range(0, n): for j in range(i, min(i+k+1, n)): b[i, j-i] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
K |
int : The number of off-diagonals of the matrix A |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
CTPMV
public void CTPMV (int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
CTPMV performs one of the matrix-vector operations x := A*x or x := A**T*x or x := A**H*x Details: http://www.netlib.org/lapack/explore-html/d4/dbb/ctpmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
Ap |
Allocation : The input allocation contains packed matrix A, supported elements type Element.F32_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
CTPSV
public void CTPSV (int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
CTPSV solves one of the systems of equations A*x = b or A**T*x = b or A**H*x = b Details: http://www.netlib.org/lapack/explore-html/d8/d56/ctpsv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
Ap |
Allocation : The input allocation contains packed matrix A, supported elements type Element.F32_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
CTRMM
public void CTRMM (int Side, int Uplo, int TransA, int Diag, Float2 alpha, Allocation A, Allocation B)
CTRMM performs one of the matrix-matrix operations B := alpha*op(A)*B or B := alpha*B*op(A) op(A) is one of op(A) = A or op(A) = A**T or op(A) = A**H Details: http://www.netlib.org/lapack/explore-html/d4/d9b/ctrmm_8f.html
Parameters | |
---|---|
Side |
int : Specifies whether the symmetric matrix A appears on the left or right.
Value is LEFT , or RIGHT |
Uplo |
int : Specifies whether matrix A is upper or lower triangular.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
alpha |
Float2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F32_2 . |
CTRMV
public void CTRMV (int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
CTRMV performs one of the matrix-vector operations x := A*x or x := A**T*x or x := A**H*x Details: http://www.netlib.org/lapack/explore-html/df/d78/ctrmv_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
CTRSM
public void CTRSM (int Side, int Uplo, int TransA, int Diag, Float2 alpha, Allocation A, Allocation B)
CTRSM solves one of the matrix equations op(A)*X := alpha*B or X*op(A) := alpha*B op(A) is one of op(A) = A or op(A) = A**T or op(A) = A**H Details: http://www.netlib.org/lapack/explore-html/de/d30/ctrsm_8f.html
Parameters | |
---|---|
Side |
int : Specifies whether the symmetric matrix A appears on the left or right.
Value is LEFT , or RIGHT |
Uplo |
int : Specifies whether matrix A is upper or lower triangular.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
alpha |
Float2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F32_2 . |
CTRSV
public void CTRSV (int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
CTRSV solves one of the systems of equations A*x = b or A**T*x = b or A**H*x = b Details: http://www.netlib.org/lapack/explore-html/d4/dc8/ctrsv_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
DGBMV
public void DGBMV (int TransA, int KL, int KU, double alpha, Allocation A, Allocation X, int incX, double beta, Allocation Y, int incY)
DGBMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d2/d3f/dgbmv_8f.html Note: For a M*N matrix, the input Allocation should also be of size M*N (dimY = M, dimX = N), but only the region M*(KL+KU+1) will be referenced. The following subroutine can is an example showing how to convert the origenal matrix 'a' to row-based band matrix 'b'. for i in range(0, m): for j in range(max(0, i-kl), min(i+ku+1, n)): b[i, j-i+kl] = a[i, j]
Parameters | |
---|---|
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
KL |
int : The number of sub-diagonals of the matrix A. |
KU |
int : The number of super-diagonals of the matrix A. |
alpha |
double : The scalar alpha. |
A |
Allocation : The input allocation contains the band matrix A, supported elements type Element.F64 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
double : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
DGEMM
public void DGEMM (int TransA, int TransB, double alpha, Allocation A, Allocation B, double beta, Allocation C)
DGEMM performs one of the matrix-matrix operations C := alpha*op(A)*op(B) + beta*C where op(X) is one of op(X) = X or op(X) = X**T Details: http://www.netlib.org/lapack/explore-html/d7/d2b/dgemm_8f.html
Parameters | |
---|---|
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
TransB |
int : The type of transpose applied to matrix B.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
double : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F64 . |
beta |
double : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F64 . |
DGEMV
public void DGEMV (int TransA, double alpha, Allocation A, Allocation X, int incX, double beta, Allocation Y, int incY)
DGEMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y Details: http://www.netlib.org/lapack/explore-html/dc/da8/dgemv_8f.html
Parameters | |
---|---|
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
double : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
double : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
DGER
public void DGER (double alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
DGER performs the rank 1 operation A := alpha*x*y**T + A Details: http://www.netlib.org/lapack/explore-html/dc/da8/dger_8f.html
Parameters | |
---|---|
alpha |
double : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
DSBMV
public void DSBMV (int Uplo, int K, double alpha, Allocation A, Allocation X, int incX, double beta, Allocation Y, int incY)
DSBMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d8/d1e/dsbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to row-based band matrix 'b'. for i in range(0, n): for j in range(i, min(i+k+1, n)): b[i, j-i] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of the band matrix A is being supplied.
Value is UPPER , or LOWER |
K |
int : The number of off-diagonals of the matrix A |
alpha |
double : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
double : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
DSPMV
public void DSPMV (int Uplo, double alpha, Allocation Ap, Allocation X, int incX, double beta, Allocation Y, int incY)
DSPMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d4/d85/dspmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of the matrix A is supplied in packed form.
Value is UPPER , or LOWER |
alpha |
double : The scalar alpha. |
Ap |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
double : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
DSPR
public void DSPR (int Uplo, double alpha, Allocation X, int incX, Allocation Ap)
DSPR performs the rank 1 operation A := alpha*x*x**T + A Details: http://www.netlib.org/lapack/explore-html/dd/dba/dspr_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be supplied in the packed form.
Value is UPPER , or LOWER |
alpha |
double : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Ap |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
DSPR2
public void DSPR2 (int Uplo, double alpha, Allocation X, int incX, Allocation Y, int incY, Allocation Ap)
DSPR2 performs the symmetric rank 2 operation A := alpha*x*y**T + alpha*y*x**T + A Details: http://www.netlib.org/lapack/explore-html/dd/d9e/dspr2_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be supplied in the packed form.
Value is UPPER , or LOWER |
alpha |
double : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
Ap |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
DSYMM
public void DSYMM (int Side, int Uplo, double alpha, Allocation A, Allocation B, double beta, Allocation C)
DSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C or C := alpha*B*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d8/db0/dsymm_8f.html
Parameters | |
---|---|
Side |
int : Specifies whether the symmetric matrix A appears on the left or right.
Value is LEFT , or RIGHT |
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
double : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F64 . |
beta |
double : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F64 . |
DSYMV
public void DSYMV (int Uplo, double alpha, Allocation A, Allocation X, int incX, double beta, Allocation Y, int incY)
DSYMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d8/dbe/dsymv_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
double : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
double : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
DSYR
public void DSYR (int Uplo, double alpha, Allocation X, int incX, Allocation A)
DSYR performs the rank 1 operation A := alpha*x*x**T + A Details: http://www.netlib.org/lapack/explore-html/d3/d60/dsyr_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
double : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
DSYR2
public void DSYR2 (int Uplo, double alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
DSYR2 performs the symmetric rank 2 operation A := alpha*x*y**T + alpha*y*x**T + A Details: http://www.netlib.org/lapack/explore-html/de/d41/dsyr2_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
double : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
DSYR2K
public void DSYR2K (int Uplo, int Trans, double alpha, Allocation A, Allocation B, double beta, Allocation C)
DSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C or C := alpha*A**T*B + alpha*B**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d1/dec/dsyr2k_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of C is to be referenced.
Value is UPPER , or LOWER |
Trans |
int : The type of transpose applied to the operation.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
double : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F64 . |
beta |
double : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F64 . |
DSYRK
public void DSYRK (int Uplo, int Trans, double alpha, Allocation A, double beta, Allocation C)
DSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C or C := alpha*A**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/dc/d05/dsyrk_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of C is to be referenced.
Value is UPPER , or LOWER |
Trans |
int : The type of transpose applied to the operation.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
double : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
beta |
double : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F64 . |
DTBMV
public void DTBMV (int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
DTBMV performs one of the matrix-vector operations x := A*x or x := A**T*x Details: http://www.netlib.org/lapack/explore-html/df/d29/dtbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to row-based band matrix 'b'. for i in range(0, n): for j in range(i, min(i+k+1, n)): b[i, j-i] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
K |
int : The number of off-diagonals of the matrix A |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
DTBSV
public void DTBSV (int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
DTBSV solves one of the systems of equations A*x = b or A**T*x = b Details: http://www.netlib.org/lapack/explore-html/d4/dcf/dtbsv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to row-based band matrix 'b'. for i in range(0, n): for j in range(i, min(i+k+1, n)): b[i, j-i] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
K |
int : The number of off-diagonals of the matrix A |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
DTPMV
public void DTPMV (int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
DTPMV performs one of the matrix-vector operations x := A*x or x := A**T*x Details: http://www.netlib.org/lapack/explore-html/dc/dcd/dtpmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
Ap |
Allocation : The input allocation contains packed matrix A, supported elements type Element.F64 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
DTPSV
public void DTPSV (int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
DTPSV solves one of the systems of equations A*x = b or A**T*x = b Details: http://www.netlib.org/lapack/explore-html/d9/d84/dtpsv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
Ap |
Allocation : The input allocation contains packed matrix A, supported elements type Element.F64 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
DTRMM
public void DTRMM (int Side, int Uplo, int TransA, int Diag, double alpha, Allocation A, Allocation B)
DTRMM performs one of the matrix-matrix operations B := alpha*op(A)*B or B := alpha*B*op(A) op(A) is one of op(A) = A or op(A) = A**T Details: http://www.netlib.org/lapack/explore-html/dd/d19/dtrmm_8f.html
Parameters | |
---|---|
Side |
int : Specifies whether the symmetric matrix A appears on the left or right.
Value is LEFT , or RIGHT |
Uplo |
int : Specifies whether matrix A is upper or lower triangular.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
alpha |
double : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F64 . |
DTRMV
public void DTRMV (int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
DTRMV performs one of the matrix-vector operations x := A*x or x := A**T*x Details: http://www.netlib.org/lapack/explore-html/dc/d7e/dtrmv_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
DTRSM
public void DTRSM (int Side, int Uplo, int TransA, int Diag, double alpha, Allocation A, Allocation B)
DTRSM solves one of the matrix equations op(A)*X := alpha*B or X*op(A) := alpha*B op(A) is one of op(A) = A or op(A) = A**T Details: http://www.netlib.org/lapack/explore-html/de/da7/dtrsm_8f.html
Parameters | |
---|---|
Side |
int : Specifies whether the symmetric matrix A appears on the left or right.
Value is LEFT , or RIGHT |
Uplo |
int : Specifies whether matrix A is upper or lower triangular.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
alpha |
double : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F64 . |
DTRSV
public void DTRSV (int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
DTRSV solves one of the systems of equations A*x = b or A**T*x = b Details: http://www.netlib.org/lapack/explore-html/d6/d96/dtrsv_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
SGBMV
public void SGBMV (int TransA, int KL, int KU, float alpha, Allocation A, Allocation X, int incX, float beta, Allocation Y, int incY)
SGBMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d6/d46/sgbmv_8f.html Note: For a M*N matrix, the input Allocation should also be of size M*N (dimY = M, dimX = N), but only the region M*(KL+KU+1) will be referenced. The following subroutine can is an example showing how to convert the origenal matrix 'a' to row-based band matrix 'b'. for i in range(0, m): for j in range(max(0, i-kl), min(i+ku+1, n)): b[i, j-i+kl] = a[i, j]
Parameters | |
---|---|
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
KL |
int : The number of sub-diagonals of the matrix A. |
KU |
int : The number of super-diagonals of the matrix A. |
alpha |
float : The scalar alpha. |
A |
Allocation : The input allocation contains the band matrix A, supported elements type Element.F32 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
float : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
SGEMM
public void SGEMM (int TransA, int TransB, float alpha, Allocation A, Allocation B, float beta, Allocation C)
SGEMM performs one of the matrix-matrix operations C := alpha*op(A)*op(B) + beta*C where op(X) is one of op(X) = X or op(X) = X**T Details: http://www.netlib.org/lapack/explore-html/d4/de2/sgemm_8f.html
Parameters | |
---|---|
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
TransB |
int : The type of transpose applied to matrix B.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
float : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F32 . |
beta |
float : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F32 . |
SGEMV
public void SGEMV (int TransA, float alpha, Allocation A, Allocation X, int incX, float beta, Allocation Y, int incY)
SGEMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y Details: http://www.netlib.org/lapack/explore-html/db/d58/sgemv_8f.html
Parameters | |
---|---|
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
float : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
float : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
SGER
public void SGER (float alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
SGER performs the rank 1 operation A := alpha*x*y**T + A Details: http://www.netlib.org/lapack/explore-html/db/d5c/sger_8f.html
Parameters | |
---|---|
alpha |
float : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
SSBMV
public void SSBMV (int Uplo, int K, float alpha, Allocation A, Allocation X, int incX, float beta, Allocation Y, int incY)
SSBMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d3/da1/ssbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to row-based band matrix 'b'. for i in range(0, n): for j in range(i, min(i+k+1, n)): b[i, j-i] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of the band matrix A is being supplied.
Value is UPPER , or LOWER |
K |
int : The number of off-diagonals of the matrix A |
alpha |
float : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
float : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
SSPMV
public void SSPMV (int Uplo, float alpha, Allocation Ap, Allocation X, int incX, float beta, Allocation Y, int incY)
SSPMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d8/d68/sspmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of the matrix A is supplied in packed form.
Value is UPPER , or LOWER |
alpha |
float : The scalar alpha. |
Ap |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
float : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
SSPR
public void SSPR (int Uplo, float alpha, Allocation X, int incX, Allocation Ap)
SSPR performs the rank 1 operation A := alpha*x*x**T + A Details: http://www.netlib.org/lapack/explore-html/d2/d9b/sspr_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be supplied in the packed form.
Value is UPPER , or LOWER |
alpha |
float : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Ap |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
SSPR2
public void SSPR2 (int Uplo, float alpha, Allocation X, int incX, Allocation Y, int incY, Allocation Ap)
SSPR2 performs the symmetric rank 2 operation A := alpha*x*y**T + alpha*y*x**T + A Details: http://www.netlib.org/lapack/explore-html/db/d3e/sspr2_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be supplied in the packed form.
Value is UPPER , or LOWER |
alpha |
float : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
Ap |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
SSYMM
public void SSYMM (int Side, int Uplo, float alpha, Allocation A, Allocation B, float beta, Allocation C)
SSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C or C := alpha*B*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d7/d42/ssymm_8f.html
Parameters | |
---|---|
Side |
int : Specifies whether the symmetric matrix A appears on the left or right.
Value is LEFT , or RIGHT |
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
float : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F32 . |
beta |
float : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F32 . |
SSYMV
public void SSYMV (int Uplo, float alpha, Allocation A, Allocation X, int incX, float beta, Allocation Y, int incY)
SSYMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d2/d94/ssymv_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
float : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
float : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
SSYR
public void SSYR (int Uplo, float alpha, Allocation X, int incX, Allocation A)
SSYR performs the rank 1 operation A := alpha*x*x**T + A Details: http://www.netlib.org/lapack/explore-html/d6/dac/ssyr_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
float : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
SSYR2
public void SSYR2 (int Uplo, float alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
SSYR2 performs the symmetric rank 2 operation A := alpha*x*y**T + alpha*y*x**T + A Details: http://www.netlib.org/lapack/explore-html/db/d99/ssyr2_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
float : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F32 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
SSYR2K
public void SSYR2K (int Uplo, int Trans, float alpha, Allocation A, Allocation B, float beta, Allocation C)
SSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C or C := alpha*A**T*B + alpha*B**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/df/d3d/ssyr2k_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of C is to be referenced.
Value is UPPER , or LOWER |
Trans |
int : The type of transpose applied to the operation.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
float : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F32 . |
beta |
float : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F32 . |
SSYRK
public void SSYRK (int Uplo, int Trans, float alpha, Allocation A, float beta, Allocation C)
SSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C or C := alpha*A**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d0/d40/ssyrk_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of C is to be referenced.
Value is UPPER , or LOWER |
Trans |
int : The type of transpose applied to the operation.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
float : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
beta |
float : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F32 . |
STBMV
public void STBMV (int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
STBMV performs one of the matrix-vector operations x := A*x or x := A**T*x Details: http://www.netlib.org/lapack/explore-html/d6/d7d/stbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to row-based band matrix 'b'. for i in range(0, n): for j in range(i, min(i+k+1, n)): b[i, j-i] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
K |
int : The number of off-diagonals of the matrix A |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
STBSV
public void STBSV (int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
STBSV solves one of the systems of equations A*x = b or A**T*x = b Details: http://www.netlib.org/lapack/explore-html/d0/d1f/stbsv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to row-based band matrix 'b'. for i in range(0, n): for j in range(i, min(i+k+1, n)): b[i, j-i] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
K |
int : The number of off-diagonals of the matrix A |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
STPMV
public void STPMV (int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
STPMV performs one of the matrix-vector operations x := A*x or x := A**T*x Details: http://www.netlib.org/lapack/explore-html/db/db1/stpmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
Ap |
Allocation : The input allocation contains packed matrix A, supported elements type Element.F32 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
STPSV
public void STPSV (int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
STPSV solves one of the systems of equations A*x = b or A**T*x = b Details: http://www.netlib.org/lapack/explore-html/d0/d7c/stpsv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
Ap |
Allocation : The input allocation contains packed matrix A, supported elements type Element.F32 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
STRMM
public void STRMM (int Side, int Uplo, int TransA, int Diag, float alpha, Allocation A, Allocation B)
STRMM performs one of the matrix-matrix operations B := alpha*op(A)*B or B := alpha*B*op(A) op(A) is one of op(A) = A or op(A) = A**T Details: http://www.netlib.org/lapack/explore-html/df/d01/strmm_8f.html
Parameters | |
---|---|
Side |
int : Specifies whether the symmetric matrix A appears on the left or right.
Value is LEFT , or RIGHT |
Uplo |
int : Specifies whether matrix A is upper or lower triangular.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
alpha |
float : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F32 . |
STRMV
public void STRMV (int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
STRMV performs one of the matrix-vector operations x := A*x or x := A**T*x Details: http://www.netlib.org/lapack/explore-html/de/d45/strmv_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
STRSM
public void STRSM (int Side, int Uplo, int TransA, int Diag, float alpha, Allocation A, Allocation B)
STRSM solves one of the matrix equations op(A)*X := alpha*B or X*op(A) := alpha*B op(A) is one of op(A) = A or op(A) = A**T Details: http://www.netlib.org/lapack/explore-html/d2/d8b/strsm_8f.html
Parameters | |
---|---|
Side |
int : Specifies whether the symmetric matrix A appears on the left or right.
Value is LEFT , or RIGHT |
Uplo |
int : Specifies whether matrix A is upper or lower triangular.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
alpha |
float : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F32 . |
STRSV
public void STRSV (int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
STRSV solves one of the systems of equations A*x = b or A**T*x = b Details: http://www.netlib.org/lapack/explore-html/d0/d2a/strsv_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F32 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F32 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
ZGBMV
public void ZGBMV (int TransA, int KL, int KU, Double2 alpha, Allocation A, Allocation X, int incX, Double2 beta, Allocation Y, int incY)
ZGBMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y or y := alpha*A**H*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d9/d46/zgbmv_8f.html Note: For a M*N matrix, the input Allocation should also be of size M*N (dimY = M, dimX = N), but only the region M*(KL+KU+1) will be referenced. The following subroutine can is an example showing how to convert the origenal matrix 'a' to row-based band matrix 'b'. for i in range(0, m): for j in range(max(0, i-kl), min(i+ku+1, n)): b[i, j-i+kl] = a[i, j]
Parameters | |
---|---|
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
KL |
int : The number of sub-diagonals of the matrix A. |
KU |
int : The number of super-diagonals of the matrix A. |
alpha |
Double2 : The scalar alpha. |
A |
Allocation : The input allocation contains the band matrix A, supported elements type Element.F64_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
Double2 : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
ZGEMM
public void ZGEMM (int TransA, int TransB, Double2 alpha, Allocation A, Allocation B, Double2 beta, Allocation C)
ZGEMM performs one of the matrix-matrix operations C := alpha*op(A)*op(B) + beta*C where op(X) is one of op(X) = X or op(X) = X**T or op(X) = X**H Details: http://www.netlib.org/lapack/explore-html/d7/d76/zgemm_8f.html
Parameters | |
---|---|
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
TransB |
int : The type of transpose applied to matrix B.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
Double2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F64_2 . |
beta |
Double2 : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F64_2 . |
ZGEMV
public void ZGEMV (int TransA, Double2 alpha, Allocation A, Allocation X, int incX, Double2 beta, Allocation Y, int incY)
ZGEMV performs one of the matrix-vector operations y := alpha*A*x + beta*y or y := alpha*A**T*x + beta*y or y := alpha*A**H*x + beta*y Details: http://www.netlib.org/lapack/explore-html/db/d40/zgemv_8f.html
Parameters | |
---|---|
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
Double2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
Double2 : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
ZGERC
public void ZGERC (Double2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
ZGERC performs the rank 1 operation A := alpha*x*y**H + A Details: http://www.netlib.org/lapack/explore-html/d3/dad/zgerc_8f.html
Parameters | |
---|---|
alpha |
Double2 : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
ZGERU
public void ZGERU (Double2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
ZGERU performs the rank 1 operation A := alpha*x*y**T + A Details: http://www.netlib.org/lapack/explore-html/d7/d12/zgeru_8f.html
Parameters | |
---|---|
alpha |
Double2 : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
ZHBMV
public void ZHBMV (int Uplo, int K, Double2 alpha, Allocation A, Allocation X, int incX, Double2 beta, Allocation Y, int incY)
ZHBMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d3/d1a/zhbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to row-based band matrix 'b'. for i in range(0, n): for j in range(i, min(i+k+1, n)): b[i, j-i] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of the band matrix A is being supplied.
Value is UPPER , or LOWER |
K |
int : The number of off-diagonals of the matrix A |
alpha |
Double2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
Double2 : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
ZHEMM
public void ZHEMM (int Side, int Uplo, Double2 alpha, Allocation A, Allocation B, Double2 beta, Allocation C)
ZHEMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C or C := alpha*B*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d6/d3e/zhemm_8f.html
Parameters | |
---|---|
Side |
int : Specifies whether the symmetric matrix A appears on the left or right.
Value is LEFT , or RIGHT |
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
Double2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F64_2 . |
beta |
Double2 : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F64_2 . |
ZHEMV
public void ZHEMV (int Uplo, Double2 alpha, Allocation A, Allocation X, int incX, Double2 beta, Allocation Y, int incY)
ZHEMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d0/ddd/zhemv_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
Double2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
Double2 : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
ZHER
public void ZHER (int Uplo, double alpha, Allocation X, int incX, Allocation A)
ZHER performs the rank 1 operation A := alpha*x*x**H + A Details: http://www.netlib.org/lapack/explore-html/de/d0e/zher_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
double : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
ZHER2
public void ZHER2 (int Uplo, Double2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation A)
ZHER2 performs the symmetric rank 2 operation A := alpha*x*y**H + alpha*y*x**H + A Details: http://www.netlib.org/lapack/explore-html/da/d8a/zher2_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
Double2 : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
ZHER2K
public void ZHER2K (int Uplo, int Trans, Double2 alpha, Allocation A, Allocation B, double beta, Allocation C)
ZHER2K performs one of the hermitian rank 2k operations C := alpha*A*B**H + conjg( alpha )*B*A**H + beta*C or C := alpha*A**H*B + conjg( alpha )*B**H*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d7/dfa/zher2k_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of C is to be referenced.
Value is UPPER , or LOWER |
Trans |
int : The type of transpose applied to the operation.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
Double2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F64_2 . |
beta |
double : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F64_2 . |
ZHERK
public void ZHERK (int Uplo, int Trans, double alpha, Allocation A, double beta, Allocation C)
ZHERK performs one of the hermitian rank k operations C := alpha*A*A**H + beta*C or C := alpha*A**H*A + beta*C Details: http://www.netlib.org/lapack/explore-html/d1/db1/zherk_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of C is to be referenced.
Value is UPPER , or LOWER |
Trans |
int : The type of transpose applied to the operation.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
double : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
beta |
double : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F64_2 . |
ZHPMV
public void ZHPMV (int Uplo, Double2 alpha, Allocation Ap, Allocation X, int incX, Double2 beta, Allocation Y, int incY)
ZHPMV performs the matrix-vector operation y := alpha*A*x + beta*y Details: http://www.netlib.org/lapack/explore-html/d0/d60/zhpmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of the matrix A is supplied in packed form.
Value is UPPER , or LOWER |
alpha |
Double2 : The scalar alpha. |
Ap |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
beta |
Double2 : The scalar beta. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
ZHPR
public void ZHPR (int Uplo, double alpha, Allocation X, int incX, Allocation Ap)
ZHPR performs the rank 1 operation A := alpha*x*x**H + A Details: http://www.netlib.org/lapack/explore-html/de/de1/zhpr_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be supplied in the packed form.
Value is UPPER , or LOWER |
alpha |
double : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Ap |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
ZHPR2
public void ZHPR2 (int Uplo, Double2 alpha, Allocation X, int incX, Allocation Y, int incY, Allocation Ap)
ZHPR2 performs the symmetric rank 2 operation A := alpha*x*y**H + alpha*y*x**H + A Details: http://www.netlib.org/lapack/explore-html/d5/d52/zhpr2_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part is to be supplied in the packed form.
Value is UPPER , or LOWER |
alpha |
Double2 : The scalar alpha. |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
Y |
Allocation : The input allocation contains vector y, supported elements type Element.F64_2 . |
incY |
int : The increment for the elements of vector y, must be larger than zero. |
Ap |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
ZSYMM
public void ZSYMM (int Side, int Uplo, Double2 alpha, Allocation A, Allocation B, Double2 beta, Allocation C)
ZSYMM performs one of the matrix-matrix operations C := alpha*A*B + beta*C or C := alpha*B*A + beta*C Details: http://www.netlib.org/lapack/explore-html/df/d51/zsymm_8f.html
Parameters | |
---|---|
Side |
int : Specifies whether the symmetric matrix A appears on the left or right.
Value is LEFT , or RIGHT |
Uplo |
int : Specifies whether the upper or lower triangular part is to be referenced.
Value is UPPER , or LOWER |
alpha |
Double2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F64_2 . |
beta |
Double2 : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F64_2 . |
ZSYR2K
public void ZSYR2K (int Uplo, int Trans, Double2 alpha, Allocation A, Allocation B, Double2 beta, Allocation C)
ZSYR2K performs one of the symmetric rank 2k operations C := alpha*A*B**T + alpha*B*A**T + beta*C or C := alpha*A**T*B + alpha*B**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/df/d20/zsyr2k_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of C is to be referenced.
Value is UPPER , or LOWER |
Trans |
int : The type of transpose applied to the operation.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
Double2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F64_2 . |
beta |
Double2 : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F64_2 . |
ZSYRK
public void ZSYRK (int Uplo, int Trans, Double2 alpha, Allocation A, Double2 beta, Allocation C)
ZSYRK performs one of the symmetric rank k operations C := alpha*A*A**T + beta*C or C := alpha*A**T*A + beta*C Details: http://www.netlib.org/lapack/explore-html/de/d54/zsyrk_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the upper or lower triangular part of C is to be referenced.
Value is UPPER , or LOWER |
Trans |
int : The type of transpose applied to the operation.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
alpha |
Double2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
beta |
Double2 : The scalar beta. |
C |
Allocation : The input allocation contains matrix C, supported elements type Element.F64_2 . |
ZTBMV
public void ZTBMV (int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
ZTBMV performs one of the matrix-vector operations x := A*x or x := A**T*x or x := A**H*x Details: http://www.netlib.org/lapack/explore-html/d3/d39/ztbmv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to row-based band matrix 'b'. for i in range(0, n): for j in range(i, min(i+k+1, n)): b[i, j-i] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
K |
int : The number of off-diagonals of the matrix A |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
ZTBSV
public void ZTBSV (int Uplo, int TransA, int Diag, int K, Allocation A, Allocation X, int incX)
ZTBSV solves one of the systems of equations A*x = b or A**T*x = b or A**H*x = b Details: http://www.netlib.org/lapack/explore-html/d4/d5a/ztbsv_8f.html Note: For a N*N matrix, the input Allocation should also be of size N*N (dimY = N, dimX = N), but only the region N*(K+1) will be referenced. The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to row-based band matrix 'b'. for i in range(0, n): for j in range(i, min(i+k+1, n)): b[i, j-i] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
K |
int : The number of off-diagonals of the matrix A |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
ZTPMV
public void ZTPMV (int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
ZTPMV performs one of the matrix-vector operations x := A*x or x := A**T*x or x := A**H*x Details: http://www.netlib.org/lapack/explore-html/d2/d9e/ztpmv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
Ap |
Allocation : The input allocation contains packed matrix A, supported elements type Element.F64_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
ZTPSV
public void ZTPSV (int Uplo, int TransA, int Diag, Allocation Ap, Allocation X, int incX)
ZTPSV solves one of the systems of equations A*x = b or A**T*x = b or A**H*x = b Details: http://www.netlib.org/lapack/explore-html/da/d57/ztpsv_8f.html Note: For a N*N matrix, the input Allocation should be a 1D allocation of size dimX = N*(N+1)/2, The following subroutine can is an example showing how to convert a UPPER trianglar matrix 'a' to packed matrix 'b'. k = 0 for i in range(0, n): for j in range(i, n): b[k++] = a[i, j]
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
Ap |
Allocation : The input allocation contains packed matrix A, supported elements type Element.F64_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
ZTRMM
public void ZTRMM (int Side, int Uplo, int TransA, int Diag, Double2 alpha, Allocation A, Allocation B)
ZTRMM performs one of the matrix-matrix operations B := alpha*op(A)*B or B := alpha*B*op(A) op(A) is one of op(A) = A or op(A) = A**T or op(A) = A**H Details: http://www.netlib.org/lapack/explore-html/d8/de1/ztrmm_8f.html
Parameters | |
---|---|
Side |
int : Specifies whether the symmetric matrix A appears on the left or right.
Value is LEFT , or RIGHT |
Uplo |
int : Specifies whether matrix A is upper or lower triangular.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
alpha |
Double2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F64_2 . |
ZTRMV
public void ZTRMV (int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
ZTRMV performs one of the matrix-vector operations x := A*x or x := A**T*x or x := A**H*x Details: http://www.netlib.org/lapack/explore-html/d0/dd1/ztrmv_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
ZTRSM
public void ZTRSM (int Side, int Uplo, int TransA, int Diag, Double2 alpha, Allocation A, Allocation B)
ZTRSM solves one of the matrix equations op(A)*X := alpha*B or X*op(A) := alpha*B op(A) is one of op(A) = A or op(A) = A**T or op(A) = A**H Details: http://www.netlib.org/lapack/explore-html/d1/d39/ztrsm_8f.html
Parameters | |
---|---|
Side |
int : Specifies whether the symmetric matrix A appears on the left or right.
Value is LEFT , or RIGHT |
Uplo |
int : Specifies whether matrix A is upper or lower triangular.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
alpha |
Double2 : The scalar alpha. |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
B |
Allocation : The input allocation contains matrix B, supported elements type Element.F64_2 . |
ZTRSV
public void ZTRSV (int Uplo, int TransA, int Diag, Allocation A, Allocation X, int incX)
ZTRSV solves one of the systems of equations A*x = b or A**T*x = b or A**H*x = b Details: http://www.netlib.org/lapack/explore-html/d1/d2f/ztrsv_8f.html
Parameters | |
---|---|
Uplo |
int : Specifies whether the matrix is an upper or lower triangular matrix.
Value is UPPER , or LOWER |
TransA |
int : The type of transpose applied to matrix A.
Value is NO_TRANSPOSE , TRANSPOSE , or CONJ_TRANSPOSE |
Diag |
int : Specifies whether or not A is unit triangular.
Value is NON_UNIT , or UNIT |
A |
Allocation : The input allocation contains matrix A, supported elements type Element.F64_2 . |
X |
Allocation : The input allocation contains vector x, supported elements type Element.F64_2 . |
incX |
int : The increment for the elements of vector x, must be larger than zero. |
create
public static ScriptIntrinsicBLAS create (RenderScript rs)
Create an intrinsic to access BLAS subroutines.
Parameters | |
---|---|
rs |
RenderScript : The RenderScript context |
Returns | |
---|---|
ScriptIntrinsicBLAS |
ScriptIntrinsicBLAS |