seemps.analysis.cross.maxvol_square#
- seemps.analysis.cross.maxvol_square(A: NDArray, max_iter: int = 10, tol: float = 1.05) tuple[NDArray, NDArray][source]#
Returns the row indices I of a tall matrix A of size (n x r) with n > r that give place to a square submatrix of (quasi-)maximum volume (modulus of the submatrix determinant). Also, returns a matrix of coefficients B such that A ≈ B A[I, :].
- Parameters:
- A
np.ndarray A tall (n x r) matrix with more rows than columns (n > r).
- maxiter
int, default=100 Maximum number of iterations allowed.
- tol
float, default=1.1 Sensibility of the algorithm.
- A
- Returns:
- I
np.ndarray An array of r indices that determine a square submatrix of A with (quasi-)maximum volume.
- B
np.ndarray A (r x r) submatrix of coefficients such that A ≈ B A[I, :].
- I