Custom extensions of sklearn functionalities

Custom expansions of sklearn functionalities.

Note

This module provides custom expansions of some sklearn classes and functions which are necessary to fit the purposes for the desired functionalities of the MLR module. As long-term goal we would like to include these functionalities to the sklearn package since we believe these additions might be helpful for everyone. This module serves as interim solution. To ensure that all features are properly working this module is also covered by extensive tests.

Parts of this code have been copied from sklearn.

License: BSD 3-Clause License

Copyright (c) 2007-2020 The scikit-learn developers. All rights reserved.

Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:

  • Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer.

  • Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution.

  • Neither the name of the copyright holder nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS “AS IS” AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

Classes:

AdvancedPipeline(*args, **kwargs)

Expand sklearn.pipeline.Pipeline.

AdvancedRFE(*args, **kwargs)

Expand sklearn.feature_selection.RFE.

AdvancedRFECV(*args, **kwargs)

Expand sklearn.feature_selection.RFECV.

AdvancedTransformedTargetRegressor(*args, ...)

Expand sklearn.compose.TransformedTargetRegressor.

FeatureSelectionTransformer(*args, **kwargs)

Transformer step of a feature selection estimator.

Functions:

cross_val_score_weighted(estimator, x_data)

Expand sklearn.model_selection.cross_val_score().

get_rfecv_transformer(rfecv_estimator)

Get transformer step of RFECV estimator.

perform_efecv(estimator, x_data, y_data, ...)

Perform exhaustive feature selection.

class esmvaltool.diag_scripts.mlr.custom_sklearn.AdvancedPipeline(*args: Any, **kwargs: Any)[source]

Bases: sklearn.pipeline.

Expand sklearn.pipeline.Pipeline.

Attributes:

coef_

Model coefficients.

feature_importances_

Feature importances.

Methods:

fit_target_transformer_only(y_data, **fit_kwargs)

Fit only transform step of of target regressor.

fit_transformers_only(x_data, y_data, ...)

Fit only transform steps of Pipeline.

transform_only(x_data)

Only perform transform steps of Pipeline.

transform_target_only(y_data)

Only perform transform steps of target regressor.

property coef_

Model coefficients.

Type

numpy.ndarray

property feature_importances_

Feature importances.

Type

numpy.ndarray

fit_target_transformer_only(y_data, **fit_kwargs)[source]

Fit only transform step of of target regressor.

fit_transformers_only(x_data, y_data, **fit_kwargs)[source]

Fit only transform steps of Pipeline.

transform_only(x_data)[source]

Only perform transform steps of Pipeline.

transform_target_only(y_data)[source]

Only perform transform steps of target regressor.

class esmvaltool.diag_scripts.mlr.custom_sklearn.AdvancedRFE(*args: Any, **kwargs: Any)[source]

Bases: sklearn.feature_selection.

Expand sklearn.feature_selection.RFE.

Methods:

fit(x_data, y_data, **fit_kwargs)

Expand fit() to accept kwargs.

predict(*args, **kwargs)

fit(x_data, y_data, **fit_kwargs)[source]

Expand fit() to accept kwargs.

predict(x_data, **predict_kwargs)

Expand predict() to accept kwargs.

class esmvaltool.diag_scripts.mlr.custom_sklearn.AdvancedRFECV(*args: Any, **kwargs: Any)[source]

Bases: sklearn.feature_selection.

Expand sklearn.feature_selection.RFECV.

Methods:

fit(x_data, y_data[, groups])

Expand fit() to accept kwargs.

predict(*args, **kwargs)

fit(x_data, y_data, groups=None, **fit_kwargs)[source]

Expand fit() to accept kwargs.

predict(x_data, **predict_kwargs)

Expand predict() to accept kwargs.

class esmvaltool.diag_scripts.mlr.custom_sklearn.AdvancedTransformedTargetRegressor(*args: Any, **kwargs: Any)[source]

Bases: sklearn.compose.

Expand sklearn.compose.TransformedTargetRegressor.

Attributes:

coef_

Model coefficients.

feature_importances_

Feature importances.

Methods:

fit(x_data, y_data, **fit_kwargs)

Expand fit() to accept kwargs.

fit_transformer_only(y_data, **fit_kwargs)

Fit only transformer step.

predict(x_data[, always_return_1d])

Expand predict() to accept kwargs.

property coef_

Model coefficients.

Type

numpy.ndarray

property feature_importances_

Feature importances.

Type

numpy.ndarray

fit(x_data, y_data, **fit_kwargs)[source]

Expand fit() to accept kwargs.

fit_transformer_only(y_data, **fit_kwargs)[source]

Fit only transformer step.

predict(x_data, always_return_1d=True, **predict_kwargs)[source]

Expand predict() to accept kwargs.

class esmvaltool.diag_scripts.mlr.custom_sklearn.FeatureSelectionTransformer(*args: Any, **kwargs: Any)[source]

Bases: sklearn.base., sklearn.feature_selection.

Transformer step of a feature selection estimator.

Methods:

fit(*_, **__)

Empty method.

fit(*_, **__)[source]

Empty method.

esmvaltool.diag_scripts.mlr.custom_sklearn.cross_val_score_weighted(estimator, x_data, y_data=None, groups=None, scoring=None, cv=None, n_jobs=None, verbose=0, fit_params=None, pre_dispatch='2*n_jobs', error_score=nan, sample_weights=None)[source]

Expand sklearn.model_selection.cross_val_score().

esmvaltool.diag_scripts.mlr.custom_sklearn.get_rfecv_transformer(rfecv_estimator)[source]

Get transformer step of RFECV estimator.

esmvaltool.diag_scripts.mlr.custom_sklearn.perform_efecv(estimator, x_data, y_data, **kwargs)[source]

Perform exhaustive feature selection.