SubtractReference#

class chemotools.baseline.SubtractReference(reference: ndarray | None = None, scale_reference: bool = False, start: int = 0, end: int = -1, x_axis: ndarray | None = None)[source]

Bases: XAxisMixin, TransformerMixin, OneToOneFeatureMixin, BaseEstimator

Subtract a reference spectrum from spectral data.

By default, the transformer computes \(x - r\) for each sample. When scale_reference=True, the reference is first scaled by an optimal factor \(a\) that solves:

\[\min_a \|x - a \cdot r\|_2\]

and returns \(x - a \cdot r\). The factor can be computed over a sub-range of the spectrum defined by start and end.

Parameters:

  • reference (np.ndarray, optional, default=None) – The reference spectrum to subtract from the input data. If None, the original spectrum is returned.

  • scale_reference (bool, default=False) – If True, the reference is scaled by a factor \(a\) before subtraction, where \(a\) minimises \(\|x - a \cdot r\|_2\) (or over the sub-range defined by start / end). If False, a simple subtraction \(x - r\) is performed and start, end, and x_axis are ignored.

  • start (int, default=0) – Index or x-axis value of the start of the range used to compute the scaling factor. Only used when scale_reference=True.

  • end (int, default=-1) – Index or x-axis value of the end of the range used to compute the scaling factor. Only used when scale_reference=True.

  • x_axis (array-like, optional) – X-axis values corresponding to columns. When provided, start and end are interpreted as x-axis values and the closest indices are used. Must be ascending if provided.

Variables:

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

  • reference (np.ndarray or None) – The reference spectrum to subtract from the input data. None if the reference parameter was not provided.

Examples

>>> from chemotools.baseline import SubtractReference
>>> from chemotools.datasets import load_fermentation_train
>>> # Load sample data
>>> X, _ = load_fermentation_train()
>>> # Convert X to a numpy array
>>> X = np.array(X)
>>> # Instantiate the transformer with a reference spectrum
>>> reference = X[0]
>>> transformer = SubtractReference(reference=reference)
SubtractReference()
>>> transformer.fit(X)
>>> # Generate baseline-corrected data
>>> X_corrected = transformer.transform(X)
fit(X: ndarray, y=None) → SubtractReference[source]

Fit the transformer to the input data.

Parameters:

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

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

Returns:

self – The fitted transformer.

Return type:

SubtractReference

transform(X: ndarray, y=None) → ndarray[source]

Transform the input data by subtracting the reference spectrum.

Parameters:

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

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

Returns:

X_transformed – The transformed data.

Return type:

np.ndarray of shape (n_samples, n_features)