redback.analysis.PhotometricClassifier

class redback.analysis.PhotometricClassifier(model_templates: list | None = None)

Bases: object

Classify transients from light curve shape using redback photometric models.

Compares an observed normalised light curve against a set of representative model light curves using dynamic time warping (DTW), which is robust to 10–20 day timing offsets between objects of the same type.

Returns a ClassificationResult with method=’photometric’.

__init__(model_templates: list | None = None) → None

Parameters:

model_templates – List of (model_name, parameters_dict, label) tuples. If None, uses built-in defaults.

__call__(**kwargs)

Call self as a function.

Methods

__init__([model_templates])
classify_from_lightcurve(transient[, ...])	Classify a transient from its bolometric or single-band light curve shape.

classify_from_lightcurve(transient, time_grid: ndarray | None = None, top_n: int = 5) → ClassificationResult

Classify a transient from its bolometric or single-band light curve shape.

Parameters:

• transient – A redback transient object with time and a flux/ luminosity attribute, or any object with time and flux_density arrays.

• time_grid – Time grid (days) on which to evaluate models. If None, uses the transient’s own time array.

• top_n – Number of top matches to use for probability estimation.

Returns:

ClassificationResult with method=’photometric’.