gen_poly_features#

fastcan.narx.gen_poly_features(X, ids, skip_indices=None)#

Generator to make polynomial features.

Added in version 0.5.1.

Parameters:

  • X (array-like of shape (n_samples, n_features)) – The data to transform, column by column.

  • ids (array-like of shape (n_outputs, degree)) – The unique id numbers of output features, which are formed of non-negative int values.

  • skip_indices (array-like, default=None) – Indices of features that have already been selected and can be skipped.

Yields:

  • index (int) – The index of the yielded feature.

  • feature (ndarray of shape (n_samples,)) – A polynomial feature.

Examples

>>> import numpy as np
>>> from fastcan.narx import gen_poly_features, make_poly_ids
>>> ids = make_poly_ids(2, 2)
>>> X = np.array([[1, 2], [3, 4], [5, 6], [7, 8]])
>>> for i, feat in gen_poly_features(X, ids):
...     print(i, feat)
0 [1. 3. 5. 7.]
1 [2. 4. 6. 8.]
2 [ 1.  9. 25. 49.]
3 [ 2. 12. 30. 56.]
4 [ 4. 16. 36. 64.]