LowRank.AlignDecompositions

LowRank.AlignDecompositions#

LowRank.AlignDecompositions.py

Utility for aligning scattering profiles from two decompositions onto a common q-grid. Exported at the top-level as molass.align_decompositions.

get_P_at(decomp, q_target, normalize=False)#

Return the XR scattering matrix P interpolated onto q_target.

Parameters:

  • decomp (Decomposition) – A decomposition object returned by quick_decomposition().

  • q_target (array-like, shape (m,)) – Target q-values in Å⁻¹. Must lie within the q-range of decomp.

  • normalize (bool, optional) – If True, each component column is divided by its maximum value so that all columns peak at 1. Default False.

Returns:

P_interp – Scattering matrix P evaluated at q_target.

Return type:

np.ndarray, shape (m, n_components)

align_decompositions(decomp_a, decomp_b, n=500, normalize=False)#

Interpolate two decompositions’ XR scattering matrices onto a shared q-grid.

The shared grid spans the intersection of the two q-ranges and contains n evenly-spaced points.

Parameters:

  • decomp_a (Decomposition) – Two decomposition objects to align.

  • decomp_b (Decomposition) – Two decomposition objects to align.

  • n (int, optional) – Number of q-points in the common grid. Default 500.

  • normalize (bool, optional) – If True, each component column in both matrices is normalised to peak at 1. Useful for shape comparison regardless of absolute scale. Default False.

Returns:

  • q_common (np.ndarray, shape (n,)) – The shared q-grid in Å⁻¹.

  • P_a (np.ndarray, shape (n, n_components_a)) – Scattering matrix of decomp_a evaluated on q_common.

  • P_b (np.ndarray, shape (n, n_components_b)) – Scattering matrix of decomp_b evaluated on q_common.

Examples

Compare the first scattering component of two decompositions:

from scipy.stats import pearsonr
import molass

q, P_a, P_b = molass.align_decompositions(decomp_1, decomp_2, normalize=True)
r, _ = pearsonr(P_a[:, 0], P_b[:, 0])
print(f"Pearson r = {r:.5f}")
Contents