scipy.linalg.

get_blas_funcs#

scipy.linalg.get_blas_funcs(names, arrays=(), dtype=None, ilp64='preferred')[source]#

Return available BLAS function objects from names.

Arrays are used to determine the optimal prefix of BLAS routines.

Parameters:

namesstr or sequence of str: Name(s) of BLAS functions without type prefix.
arrayssequence of ndarrays, optional: Arrays can be given to determine optimal prefix of BLAS routines. If not given, double-precision routines will be used, otherwise the most generic type in arrays will be used.
dtypestr or dtype, optional: Data-type specifier. Not used if arrays is non-empty.
ilp64{True, False, ‘preferred’}, optional: Whether to return ILP64 routine variant. Choosing 'preferred' returns ILP64 routine if available, and otherwise the 32-bit (LP64) routine. Default: 'preferred'.

Returns:

funcslist: List containing the found function(s).

Raises:

RuntimeError: If the requested LP64/ILP64 variant is not available.

See also

scipy.linalg.lapack.get_lapack_funcs: a similar routine for selecting LAPACK functions.

Notes

In BLAS, the naming convention is that all functions start with a type prefix, which depends on the type of the principal matrix. These can be one of {'s', 'd', 'c', 'z'} for the NumPy types {float32, float64, complex64, complex128} respectively. The code and the dtype are stored in attributes typecode and dtype of the returned functions.

Examples

>>> import numpy as np
>>> import scipy.linalg as LA
>>> rng = np.random.default_rng()
>>> a = rng.random((3, 2))
>>> x_gemv = LA.get_blas_funcs('gemv', (a,))

Note that x_gemv string representation shows the exact BLAS function with the prefix (here, d- because a is double precision real):

>>> x_gemv
<fortran function dgemv>

The BLAS variant information is also available from the typecode attribute:

>>> x_gemv.typecode
'd'

For double precision complex arrays, we select the z- variant, zgemv

>>> x_gemv = LA.get_blas_funcs('gemv', (a*1j,))
>>> x_gemv.typecode
'z'

If you want to select a specific BLAS variant instead of relying on array types, use the dtype= argument:

>>> LA.get_blas_funcs('gemv', dtype=np.float32)
<fortran function sgemv>

The int_dtype attribute stores whether the routine is ILP64 (integer arguments and outputs are 64-bit) or LP64 (integer arguments and outputs are 32-bit):

>>> x_gemv.int_dtype
dtype('int32')   # may vary