scikit-fallback 0.0.0

Machine learning with a reject option

scikit-fallback is a scikit-learn-compatible Python package for machine learning with a reject option.

Get started w/ scikit-fallback

Usage

from skfb.estimators import RateFallbackClassifierCV
from skfb.metrics import predict_reject_accuracy_score
from sklearn.linear_model import LogisticRegression

rejector = RateFallbackClassifierCV(
    LogisticRegression(),
    fallback_rates=[0.05, 0.07],
    cv=5,
)
rejector.fit(X_train, y_train)
y_pred = rejector.predict(X_test)
print(predict_reject_accuracy_score(y_test, y_pred))

Installation

scikit-fallback requires:

• Python (>=3.9,< 3.13)
• scikit-learn (>=1.3)

Examples

See the examples/ directory for various applications of fallback estimators and scorers to scikit-learn-compatible pipelines.