scipy.linalg.lapack.spbcon

scipy.linalg.lapack.spbcon(kd, ab, anorm, ldab=kd + 1, uplo='U') = <flapack function spbcon>

Estimate the reciprocal condition number of a positive definite band matrix from its Cholesky factorization (LAPACK spbcon).

Parameters:

kdint

Number of super- or subdiagonals of the matrix.

abndarray

Banded Cholesky factor of shape (kd + 1, n), as returned by pbtrf; the order n is read off ab itself.

anormfloat

1-norm of the original matrix, which equals its infinity norm here.

ldabint, optional

Leading dimension of ab. It must equal ab.shape[0] and be at least kd + 1, which is also the default.

uplostr, optional

'U' if ab holds the upper triangular factor, 'L' if lower. Default is 'U'.

Returns:

rcondfloat

Estimate of the reciprocal condition number.

infoint

0 on success; if negative, the -info-th argument had an illegal value.