Skip to content

operator#

Includes the contour integral kernels.

Functions:

  • operator_inverse –

    Computes the inverse of a matrix polynomial at sample points.

operator_inverse #

operator_inverse(a_xx: tuple[NDArray, ...], z: NDArray, contour_type: dtype, in_type: dtype, num_threads: int = 1024) -> NDArray

Computes the inverse of a matrix polynomial at sample points.

Parameters:

  • a_xx (tuple[NDArray, ...]) –

    The coefficients of the matrix polynomial.

  • z (NDArray) –

    The sample points at which to compute the inverse.

  • contour_type (dtype) –

    The data type for the contour integration.

  • in_type (dtype) –

    The data type for the input matrices.

  • num_threads (int, default: 1024 ) –

    The number of cuda threads to use for the kernel. Only relevant for GPU computations.

Returns:

  • NDArray –

    The inverse of the matrix polynomial.