Changelog

0.6.0 (2024-06-**)

• Added scoring parameter to metalearners.metalearner.MetaLearner.evaluate() and implemented the abstract method for the metalearners.XLearner and metalearners.DRLearner.

0.5.0 (2024-06-18)

• No longer raise an error if feature_set is provided to metalearners.SLearner.

• Fix a bug where base model dictionaries – e.g. n_folds or feature-set – were improperly initialized if the provided dictionary’s keys were a strict superset of the expected keys.

0.4.2 (2024-06-18)

• Ship license file.

0.4.1 (2024-06-18)

• Fix dependencies for pip.

0.4.0 (2024-06-18)

• Implemented metalearners.cross_fit_estimator.CrossFitEstimator.clone().

• Added n_jobs_base_learners to metalearners.metalearner.MetaLearner.fit().

• Renamed metalearners.explainer.Explainer.feature_importances(). Note this is a breaking change.

• Renamed metalearners.metalearner.MetaLearner.feature_importances(). Note this is a breaking change.

• Renamed metalearners.explainer.Explainer.shap_values(). Note this is a breaking change.

• Renamed metalearners.metalearner.MetaLearner.shap_values(). Note this is a breaking change.

• Renamed metalearners.metalearner.MetaLearner.explainer(). Note this is a breaking change.

• Implemented synchronize_cross_fitting parameter for metalearners.metalearner.MetaLearner.fit().

• Implemented cv parameter for metalearners.cross_fit_estimator.fit().

0.3.0 (2024-06-03)

• Implemented metalearners.explainer.Explainer with support for binary classification and regression outcomes and discrete treatment variants.

• Integration of metalearners.explainer.Explainer with metalearners.metalearner.MetaLearner for feature importance and SHAP values calculations.

• Implemented model reusage through the fitted_nuisance_models and fitted_propensity_model parameters of metalearners.metalearner.MetaLearner.

• Allow for fit_params in metalearners.metalearner.MetaLearner.fit().

0.2.0 (2024-05-28)

Beta release with

• metalearners.DRLearner with support for binary classification and regression outcomes and discrete treatment variants.

• Generalization of metalearners.TLearner, metalearners.XLearner and metalearners.RLearner to allow for more than two discrete treatment variants.

• Unification of shapes returned by predict methods.

• metalearners.utils.simplify_output() and metalearners.utils.metalearner_factory().

0.1.0 (2024-05-16)

Alpha release with

• metalearners.TLearner with support for binary classification and regression outcomes and binary treatment variants.

• metalearners.SLearner with support for binary classification and regression outcomes and discrete treatment variants.

• metalearners.XLearner with support for binary classification and regression outcomes and binary treatment variants.

• metalearners.RLearner with support for binary classification and regression otucomes and binary treatment variants.