RubberbandCorrection

class chemotools.baseline.RubberbandCorrection

Bases: DocLinkMixin, TransformerMixin, OneToOneFeatureMixin, BaseEstimator

A transformer that removes a baseline using the rubberband method.

The rubberband baseline is the lower convex hull of the spectrum — the set of straight-line segments a rubber band would form if stretched along the underside of the spectrum. The baseline is subtracted from each spectrum, leaving the peaks resting on a flat zero background.

The lower convex hull is computed with Andrew’s monotone chain algorithm [1] [2] in feature-index space, so the feature axis (e.g. wavenumbers) is assumed to be sorted; even spacing is not required. The method has no parameters.

Variables:

n_features_in (int) – The number of features in the input data.

Referencias

[1]

A. M. Andrew, «Another efficient algorithm for convex hulls in two dimensions», Information Processing Letters, 9(5), 216-219, 1979.

[2]

Monotone chain convex hull, reference implementation: https://en.wikibooks.org/wiki/Algorithm_Implementation/Geometry/Convex_hull/Monotone_chain

Ejemplos

>>> from chemotools.baseline import RubberbandCorrection
>>> from chemotools.datasets import load_fermentation_train
>>> # Load sample data
>>> X, _ = load_fermentation_train()
>>> # Instantiate the transformer
>>> transformer = RubberbandCorrection()
RubberbandCorrection()
>>> transformer.fit(X)
>>> # Generate baseline-corrected data
>>> X_corrected = transformer.transform(X)

fit(X: ndarray, y=None) → RubberbandCorrection

Fit the transformer to the input data.

Parámetros:

• X (np.ndarray of shape (n_samples, n_features)) – The input data to fit the transformer to.

• y (None) – Ignored to align with API.

Devuelve:

self – The fitted transformer.

Tipo del valor devuelto:

RubberbandCorrection

transform(X: ndarray, y=None) → ndarray

Transform the input data by subtracting the rubberband baseline.

Parámetros:

• X (np.ndarray of shape (n_samples, n_features)) – The input data to transform.

• y (None) – Ignored to align with API.

Devuelve:

X_transformed – The baseline-corrected data.

Tipo del valor devuelto:

np.ndarray of shape (n_samples, n_features)