scipy.linalg.interpolative.interp_decomp(A, eps_or_k, rand=True, rng=None)[source]#

Compute ID of a matrix.

An ID of a matrix A is a factorization defined by a rank k, a column index array idx, and interpolation coefficients proj such that:[:,idx[:k]], proj) = A[:,idx[k:]]

The original matrix can then be reconstructed as:

                  [:,idx[:k]], proj)]

or via the routine reconstruct_matrix_from_id. This can equivalently be written as:[:,idx[:k]],
                    numpy.hstack([numpy.eye(k), proj])

in terms of the skeleton and interpolation matrices:

B = A[:,idx[:k]]


P = numpy.hstack([numpy.eye(k), proj])[:,np.argsort(idx)]

respectively. See also reconstruct_interp_matrix and reconstruct_skel_matrix.

The ID can be computed to any relative precision or rank (depending on the value of eps_or_k). If a precision is specified (eps_or_k < 1), then this function has the output signature:

k, idx, proj = interp_decomp(A, eps_or_k)

Otherwise, if a rank is specified (eps_or_k >= 1), then the output signature is:

idx, proj = interp_decomp(A, eps_or_k)
Anumpy.ndarray or scipy.sparse.linalg.LinearOperator with rmatvec

Matrix to be factored

eps_or_kfloat or int

Relative error (if eps_or_k < 1) or rank (if eps_or_k >= 1) of approximation.

randbool, optional

Whether to use random sampling if A is of type numpy.ndarray (randomized algorithms are always used if A is of type scipy.sparse.linalg.LinearOperator).


NumPy generator for the randomization steps in the algorithm. If rand is False, the argument is ignored.


Rank required to achieve specified relative precision if eps_or_k < 1.


Column index array.


Interpolation coefficients.