Multiband features API

Multiband support in standard features

All standard feature extractors accept a bands= constructor argument to enable multiband mode. Full API documentation is available in the per-category reference pages:

• Meta features (Extractor, Bins) — Extractor supports mixed single-band + multiband, Bins supports bands= for per-passband binning.
• Periodogram — Periodogram supports bands= and multiband_normalization=.
• Variability, Linear trend, Time sampling, Non-linear parametric fits — all accept bands=.

---

Pure multiband features

These features are inherently multiband — they always require bands and have no single-band mode.

light_curve.ColorOfMaximum

Bases: _FeatureEvaluator

Difference of maximum magnitudes in two passbands.

Computes max(band[0]) - max(band[1]) where the maximum is taken over each passband independently. Note that maximum has mathematical meaning, not the astronomical one (brighter objects have smaller magnitude).

Parameters:

Name	Type	Description	Default
bands	list of two str	Two passband names. The output is m[bands[0]] - m[bands[1]].	required

Attributes:

Name	Type	Description
names	list of str	Feature names
descriptions	list of str	Feature descriptions
bands	numpy.ndarray of str or None	Passband names for multiband mode, or None for single-band mode

Methods:

Name	Description
__call__	Extract features and return them as a numpy array
many	Extract features from multiple light curves in parallel

---

light_curve.ColorOfMedian

Bases: _FeatureEvaluator

Difference of median magnitudes in two passbands.

Computes median(band[0]) - median(band[1]) where the median is taken over each passband independently.

Parameters:

Name	Type	Description	Default
bands	list of two str	Two passband names. The output is m[bands[0]] - m[bands[1]].	required

Attributes:

Name	Type	Description
names	list of str	Feature names
descriptions	list of str	Feature descriptions
bands	numpy.ndarray of str or None	Passband names for multiband mode, or None for single-band mode

Methods:

Name	Description
__call__	Extract features and return them as a numpy array
many	Extract features from multiple light curves in parallel

---

light_curve.ColorOfMinimum

Bases: _FeatureEvaluator

Difference of minimum magnitudes in two passbands.

Computes min(band[0]) - min(band[1]) where the minimum is taken over each passband independently. Note that minimum has mathematical meaning, not the astronomical one (fainter objects have larger magnitude).

Parameters:

Name	Type	Description	Default
bands	list of two str	Two passband names. The output is m[bands[0]] - m[bands[1]].	required

Attributes:

Name	Type	Description
names	list of str	Feature names
descriptions	list of str	Feature descriptions
bands	numpy.ndarray of str or None	Passband names for multiband mode, or None for single-band mode

Methods:

Name	Description
__call__	Extract features and return them as a numpy array
many	Extract features from multiple light curves in parallel

---

light_curve.ColorSpread

Bases: _FeatureEvaluator

Standard deviation of per-passband weighted mean magnitudes.

For each passband, the weighted mean magnitude is computed using inverse-variance weights. ColorSpread is then the population standard deviation of these per-band means. A large value indicates a large spread of mean brightnesses across bands; zero means all bands have the same mean magnitude.

Parameters:

Name	Type	Description	Default
bands	list of str		required
Two			required

Attributes:

Name	Type	Description
names	list of str
Feature names
descriptions	list of str
Feature descriptions
bands	numpy.ndarray of str or None
Passband names for multiband mode, or None for single-band mode

Methods:

Name	Description
__call__
Extract features and return them as a numpy array
many
Extract features from multiple light curves in parallel