Confidence Intervals

Simple evaluation of classification confidence intervals.

pip install git+https://github.com/g8a9/confidence_intervals.git

Getting Started

clf = RandomForestClassifier().fit(X_train, y_train)
y_pred = clf.predict(X_test)
bootstrap_test(y_test, y_pred)

# Output
>>> ConfidenceResult(n_iter=200, ci_level=95, mean=0.9777111111111112, ci_lower=0.9643888888888889, ci_upper=0.9888888888888889)

Features

Confidence Intervals Estimation by means of:

• bootstrap sampling on test set predictions (use n_iter > 200)
• simple estimation over multiple runs (useful for CI estimation across multiple random initializations)

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.