cupynumeric.linalg.svd#

cupynumeric.linalg.svd(
a: ndarray,
full_matrices: bool = True,
) tuple[ndarray, ...]#

Singular Value Decomposition.

Parameters:
  • a ((M, N) array_like) – Array like, at least dimension 2.

  • full_matrices (bool, optional) – If True (default), u and vh are of shape (M, M), (N, N). If False, the shapes are (M, K) and (K, N), where K = min(M, N).

Returns:

  • u ((M, M) array_like) – Unitary array(s).

  • s ((K) array_like) – The singular values, sorted in descending order

  • vh ((N, N) array_like) – Unitary array(s).

Raises:

LinAlgError – If SVD computation does not converge.

Notes

Currently does not support the parameters ‘full_matrices’, ‘compute_uv’, and ‘hermitian’.

See also

numpy.linalg.svd

Availability:

Single GPU, Single CPU