uf3.regression.regularize.combine_regularizer_matrices¶
- combine_regularizer_matrices(matrices: List) numpy.ndarray[source]¶
- Combine square penalty matrices.
- Example:
- [X— [Y- [Z– [X——–
-X– -Y] -Z- -X——- –X- –Z] –X—— —X] —X—–
- (4x4) (2x2) (3x3) -> —-Y—- (9x9)
—–Y— ——Z– ——-Z- ——–Z]
- Parameters
matrices (list) – square matrices, e.g. for separate optimization objectives like A-A, A-B, B-B interactions.
- Returns
- square penalty matrix equal in length
to the sum of constituent lengths.
- Return type
full_matrix (numpy.ndarray)