pyFM.signatures.HKS_functions.auto_HKS

pyFM.signatures.HKS_functions.auto_HKS(evals, evects, num_T, landmarks=None, scaled=True)

Compute HKS with an automatic choice of time values.

Parameters:

  • evals ((K,) np.ndarray) – The K eigenvalues.

  • evects ((N, K) np.ndarray) – The K eigenvectors.

  • num_T (int) – Number of time values to use.

  • landmarks ((p,) np.ndarray, optional) – If not None, indices of landmarks to compute.

  • scaled (bool, optional) – Whether to scale for each time value.

Returns:

HKS – Array where each column is the HKS for a given t, possibly for some landmark.

Return type:

(N, num_T) or (N, p * num_T) np.ndarray