scipy.sparse.linalg.factorized

scipy.sparse.linalg.factorized(A)[source]

Return a function for solving a sparse linear system, with A pre-factorized.

Parameters:

A : (N, N) array_like

Input.

Returns:

solve : callable

To solve the linear system of equations given in A, the solve callable should be passed an ndarray of shape (N,).

Examples

>>> from scipy.sparse.linalg import factorized
>>> A = np.array([[ 3. ,  2. , -1. ],
...               [ 2. , -2. ,  4. ],
...               [-1. ,  0.5, -1. ]])
>>> solve = factorized(A) # Makes LU decomposition.
>>> rhs1 = np.array([1, -2, 0])
>>> solve(rhs1) # Uses the LU factors.
array([ 1., -2., -2.])