scipy.linalg.lapack.cgesdd

scipy.linalg.lapack.cgesdd(a, compute_uv=1, full_matrices=1, lwork=..., overwrite_a=0) = <flapack function cgesdd>

Compute the singular value decomposition a = u @ diag(s) @ vt by divide-and-conquer (LAPACK cgesdd).

Parameters:

andarray

Matrix of shape (m, n).

compute_uvint, optional

If nonzero, singular vectors are computed. Default is 1.

full_matricesint, optional

If nonzero, full-sized u and vt are computed. Default is 1.

lworkint, optional

Size of the workspace. Default is routine-specific. Use gesdd_lwork for the optimal value.

overwrite_aint, optional

If nonzero, a may be overwritten in place. Default is 0.

Returns:

undarray

Left singular vectors; a (1, 1) placeholder when compute_uv is 0.

sndarray

Singular values of a, in descending order.

vtndarray

Right singular vectors, transposed; a placeholder when compute_uv is 0.

infoint

0 on success; if negative, the -info-th argument had an illegal value; if positive, the algorithm failed to converge.