nirs4all.operators.base.spectra_mixin module

SpectraTransformerMixin base class for wavelength-aware transformations.

This module provides a base class for spectral transformations that require wavelength information. The controller automatically provides wavelengths from the dataset when available and when the operator declares it needs them.

class nirs4all.operators.base.spectra_mixin.SpectraTransformerMixin[source]

Bases: TransformerMixin, BaseEstimator

Base class for spectral transformations that require wavelength information.

This mixin extends sklearn’s TransformerMixin to support wavelength-aware transformations. The controller automatically provides wavelengths from the dataset when available and when the operator declares it needs them.

Subclasses must implement transform_with_wavelengths() instead of transform().

Parameters:: parameters. (None - this is a mixin class. Subclasses define their own)

_requires_wavelengths

Class-level flag indicating whether this operator requires wavelengths. If True (default), transform() will raise ValueError if wavelengths are not provided. Subclasses can set this to False if wavelengths are optional.

Type:: bool

Examples

>>> class TemperatureAugmenter(SpectraTransformerMixin):
...     def __init__(self, temperature_delta: float = 5.0):
...         self.temperature_delta = temperature_delta
...
...     def transform_with_wavelengths(
...         self, X: np.ndarray, wavelengths: np.ndarray
...     ) -> np.ndarray:
...         # Apply temperature-dependent spectral changes
...         # ... implementation ...
...         return X_transformed

Notes

The controller detects SpectraTransformerMixin instances via:

needs_wavelengths = (
    isinstance(op, SpectraTransformerMixin) and
    getattr(op, '_requires_wavelengths', False)
)

Wavelengths are extracted from the dataset using dataset.wavelengths_nm(source).

fit(X, y=None, **fit_params)[source]

Fit is a no-op for most spectral transformations.

Parameters:

X (array-like of shape (n_samples, n_features)) – Training data.
y (array-like of shape (n_samples,) or None, default=None) – Target values (unused).
**fit_params (dict) – Additional fit parameters. May include ‘wavelengths’ for operators that need to fit using wavelength information.

Returns:

self – Returns self.

Return type:

object

set_transform_request(*, wavelengths: bool | None | str = '$UNCHANGED$') → SpectraTransformerMixin

Configure whether metadata should be requested to be passed to the transform method.

Note that this method is only relevant when this estimator is used as a sub-estimator within a meta-estimator and metadata routing is enabled with enable_metadata_routing=True (see sklearn.set_config()). Please check the User Guide on how the routing mechanism works.

The options for each parameter are:

True: metadata is requested, and passed to transform if provided. The request is ignored if metadata is not provided.
False: metadata is not requested and the meta-estimator will not pass it to transform.
None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.
str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Parameters:: wavelengths (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for wavelengths parameter in transform.
Returns:: self – The updated object.
Return type:: object

transform(X, wavelengths: ndarray | None = None)[source]

Transform method that delegates to transform_with_wavelengths.

If wavelengths are not provided and the operator requires them, this will raise a ValueError.

Parameters:

X (array-like of shape (n_samples, n_features)) – Input spectra array.
wavelengths (ndarray of shape (n_features,) or None, default=None) – Wavelength array in nm. Required if _requires_wavelengths is True.

Returns:

X_transformed – Transformed spectra array.

Return type:

ndarray of shape (n_samples, n_features)

Raises:

ValueError – If wavelengths are not provided and _requires_wavelengths is True.

abstractmethod transform_with_wavelengths(X: ndarray, wavelengths: ndarray | None) → ndarray[source]

Apply the transformation using wavelength information.

Subclasses must implement this method to perform the actual transformation.

Parameters:

X (ndarray of shape (n_samples, n_features)) – Input spectra.
wavelengths (ndarray of shape (n_features,) or None) – Wavelength array in nm. May be None if _requires_wavelengths is False.

Returns:

X_transformed – Transformed spectra.

Return type:

ndarray of shape (n_samples, n_features)

Raises:

NotImplementedError – If the subclass does not implement this method.