function spd_inv_split

deeptime.numeric.spd_inv_split(W, epsilon=1e-10, method='QR', canonical_signs=False)

Compute \(W^{-1} = L L^T\) of the symmetric positive-definite matrix \(W\).

by first reducing W to a low-rank approximation that is truly spd.

Parameters:

• W (ndarray((m,m), dtype=float)) – Symmetric positive-definite (spd) matrix.

• epsilon (float) – Truncation parameter. Eigenvalues with norms smaller than this cutoff will be removed.

• method (str) –

  Method to perform the decomposition of \(W\) before inverting. Options are:

  • ’QR’: QR-based robust eigenvalue decomposition of W

  • ’schur’: Schur decomposition of W

• canonical_signs (boolean, default = False) – Fix signs in L, s. t. the largest element of in every row of L is positive.

Returns:

L – Matrix \(L\) from the decomposition \(W^{-1} = L L^T\).

Return type:

ndarray((n, r))