hypernets.tabular.ensemble package

Submodules

hypernets.tabular.ensemble.base_ensemble module

class hypernets.tabular.ensemble.base_ensemble.BaseEnsemble(task, estimators, need_fit=False, n_folds=5, method='soft', random_state=9527)

Bases: object

fit(X, y, est_predictions=None)

fit_predictions(predictions, y_true)

static load(model_path)

np = <module 'numpy' from '/home/docs/checkouts/readthedocs.org/user_builds/hypernets/envs/latest/lib/python3.6/site-packages/numpy/__init__.py'>

predict(X)

predict_proba(X)

predictions2predict(predictions)

predictions2predict_proba(predictions)

proba2predict(proba, proba_threshold=0.5)

save(model_path)

hypernets.tabular.ensemble.stacking module

class hypernets.tabular.ensemble.stacking.StackingEnsemble(task, estimators, need_fit=False, n_folds=5, method='soft', meta_model=None, fit_kwargs=None)

Bases: hypernets.tabular.ensemble.base_ensemble.BaseEnsemble

fit_predictions(predictions, y_true)

predictions2predict(predictions)

predictions2predict_proba(predictions)

hypernets.tabular.ensemble.voting module

class hypernets.tabular.ensemble.voting.AveragingEnsemble(task, estimators, need_fit=False, n_folds=5, method='soft')

Bases: hypernets.tabular.ensemble.base_ensemble.BaseEnsemble

fit_predictions(predictions, y_true)

predictions2predict(predictions)

predictions2predict_proba(predictions)

class hypernets.tabular.ensemble.voting.GreedyEnsemble(task, estimators, need_fit=False, n_folds=5, method='soft', random_state=9527, scoring='neg_log_loss', ensemble_size=0)

Bases: hypernets.tabular.ensemble.base_ensemble.BaseEnsemble

References

Caruana, Rich, et al. “Ensemble selection from libraries of models.” Proceedings of the twenty-first international conference on Machine learning. 2004.

fit_predictions(predictions, y_true)

predictions2predict(predictions)

predictions2predict_proba(predictions)

Module contents