xpectra.AbstainingClassifier

class xpectra.AbstainingClassifier(model, scorer)[source]

Bases: object

Predict a class when the OOD score is below tau, otherwise abstain.

Parameters:: scorer (OODScorer)

__init__(model, scorer)[source]

Parameters:: scorer (OODScorer)
Return type:: None

Methods

`__init__`(model, scorer)
`achieved_coverage`(X)	Fraction of X that would be classified (not abstained) at the current tau.
`fit_tau`(X_val[, target_coverage])	Set tau so that target_coverage of X_val falls below it.
`predict`(X, class_names)

Attributes

ABSTAIN

ABSTAIN = 'ABSTAIN'

fit_tau(X_val, target_coverage=0.9)[source]

Set tau so that target_coverage of X_val falls below it.

X_val must be a HELD-OUT labeled set that neither the model nor the scorer was trained on — calibrating on in-sample data yields optimistic scores and a threshold too tight to hit the target coverage on new data.

Parameters:

X_val (ndarray)
target_coverage (float)

Return type:

AbstainingClassifier

achieved_coverage(X)[source]

Fraction of X that would be classified (not abstained) at the current tau.

Parameters:: X (ndarray)
Return type:: float

predict(X, class_names)[source]

Parameters:

X (ndarray)
class_names (list[str])

Return type:

tuple[ndarray, ndarray]