sparsax.umf_rcond

sparsax.umf_rcond(Ai, Aj, Ax, n)[source]

A cheap estimate of the reciprocal condition number of A, via UMFPACK.

The UMFPACK counterpart of lu_rcond(), with the same caveats: the smallest over the largest pivot, after UMFPACK’s own row scaling, so a rough indicator and not comparable with other backends. Not differentiable.

Returns:

Scalar float64 in [0, 1].

Raises:

Exception – if A is exactly singular or holds a non-finite value. set_options(on_failure="nan") returns NaN instead.