take selection docstrings out of base classes (#607)

* take docstrings out of selection base classes

* fix style

* fix indentation

* fix style

Author: solegalli

feature_engine/_docstrings/fit_attributes.py

@@ -32,3 +32,12 @@
         Dictionary with the minimum values beyond which a value will be considered an
         outlier.
     """.rstrip()
+
+# used by selection module
+_feature_importances_docstring = """feature_importances_:
+        Pandas Series with the feature importance (comes from step 2)
+    """.rstrip()
+
+_performance_drifts_docstring = """performance_drifts_:
+        Dictionary with the performance drift per examined feature (comes from step 5).
+    """.rstrip()

feature_engine/_docstrings/init_parameters/selection.py

@@ -0,0 +1,11 @@
+_confirm_variables_docstring = """confirm_variables: bool, default=False
+            If set to True, variables that are not present in the input dataframe will
+            be removed from the list of variables. Only used when passing a variable
+            list to the parameter `variables`. See parameter variables for more details.
+        """.rstrip()
+
+_estimator_docstring = """estimator: object
+            A Scikit-learn estimator for regression or classification.
+            The estimator must have either a `feature_importances` or a `coef_`
+            attribute after fitting.
+    """.rstrip()

feature_engine/selection/__init__.py

@@ -6,13 +6,13 @@
 from .drop_duplicate_features import DropDuplicateFeatures
 from .drop_features import DropFeatures
 from .drop_psi_features import DropHighPSIFeatures
+from .information_value import SelectByInformationValue
 from .recursive_feature_addition import RecursiveFeatureAddition
 from .recursive_feature_elimination import RecursiveFeatureElimination
 from .shuffle_features import SelectByShuffling
 from .single_feature_performance import SelectBySingleFeaturePerformance
 from .smart_correlation_selection import SmartCorrelatedSelection
 from .target_mean_selection import SelectByTargetMeanPerformance
-from .information_value import SelectByInformationValue
 
 __all__ = [
     "DropFeatures",

feature_engine/selection/base_recursive_selector.py

@@ -96,20 +96,6 @@ class BaseRecursiveSelector(BaseSelector):
         Find the important features.
     """
 
-    _estimator_docstring = """estimator: object
-        A Scikit-learn estimator for regression or classification.
-        The estimator must have either a `feature_importances` or a `coef_` attribute
-        after fitting.
-        """.rstrip()
-
-    _feature_importances_docstring = """feature_importances_:
-        Pandas Series with the feature importance (comes from step 2)
-        """.rstrip()
-
-    _performance_drifts_docstring = """performance_drifts_:
-        Dictionary with the performance drift per examined feature (comes from step 5).
-        """.rstrip()
-
     def __init__(
         self,
         estimator,

feature_engine/selection/base_selector.py

@@ -51,12 +51,6 @@ class BaseSelector(BaseEstimator, TransformerMixin, GetFeatureNamesOutMixin):
         Check that the variables entered by the user exist in the df.
     """
 
-    _confirm_variables_docstring = """confirm_variables: bool, default=False
-            If set to True, variables that are not present in the input dataframe will
-            be removed from the list of variables. Only used when passing a variable
-            list to the parameter `variables`. See parameter variables for more details.
-            """.rstrip()
-
     def __init__(
         self,
         confirm_variables: bool = False,

feature_engine/selection/drop_constant_features.py

@@ -6,6 +6,9 @@
     _feature_names_in_docstring,
     _n_features_in_docstring,
 )
+from feature_engine._docstrings.init_parameters.selection import (
+    _confirm_variables_docstring,
+)
 from feature_engine._docstrings.methods import _fit_transform_docstring
 from feature_engine._docstrings.substitute import Substitution
 from feature_engine._variable_handling.init_parameter_checks import (
@@ -27,7 +30,7 @@
 
 
 @Substitution(
-    confirm_variables=BaseSelector._confirm_variables_docstring,
+    confirm_variables=_confirm_variables_docstring,
     variables=_variables_all_docstring,
     variables_=_variables_attribute_docstring,
     feature_names_in_=_feature_names_in_docstring,

feature_engine/selection/drop_correlated_features.py

@@ -6,6 +6,9 @@
     _feature_names_in_docstring,
     _n_features_in_docstring,
 )
+from feature_engine._docstrings.init_parameters.selection import (
+    _confirm_variables_docstring,
+)
 from feature_engine._docstrings.methods import _fit_transform_docstring
 from feature_engine._docstrings.substitute import Substitution
 from feature_engine._variable_handling.init_parameter_checks import (
@@ -31,7 +34,7 @@
 
 
 @Substitution(
-    confirm_variables=BaseSelector._confirm_variables_docstring,
+    confirm_variables=_confirm_variables_docstring,
     variables=_variables_numerical_docstring,
     missing_values=_missing_values_docstring,
     variables_=_variables_attribute_docstring,

feature_engine/selection/drop_duplicate_features.py

@@ -6,6 +6,9 @@
     _feature_names_in_docstring,
     _n_features_in_docstring,
 )
+from feature_engine._docstrings.init_parameters.selection import (
+    _confirm_variables_docstring,
+)
 from feature_engine._docstrings.methods import _fit_transform_docstring
 from feature_engine._docstrings.substitute import Substitution
 from feature_engine._variable_handling.init_parameter_checks import (
@@ -28,7 +31,7 @@
 
 
 @Substitution(
-    confirm_variables=BaseSelector._confirm_variables_docstring,
+    confirm_variables=_confirm_variables_docstring,
     variables=_variables_all_docstring,
     missing_values=_missing_values_docstring,
     variables_=_variables_attribute_docstring,

feature_engine/selection/drop_psi_features.py

@@ -10,6 +10,9 @@
     _feature_names_in_docstring,
     _n_features_in_docstring,
 )
+from feature_engine._docstrings.init_parameters.selection import (
+    _confirm_variables_docstring,
+)
 from feature_engine._docstrings.methods import _fit_transform_docstring
 from feature_engine._docstrings.substitute import Substitution
 from feature_engine._variable_handling.init_parameter_checks import (
@@ -38,7 +41,7 @@
 
 
 @Substitution(
-    confirm_variables=BaseSelector._confirm_variables_docstring,
+    confirm_variables=_confirm_variables_docstring,
     variables=_variables_numerical_docstring,
     variables_=_variables_attribute_docstring,
     feature_names_in_=_feature_names_in_docstring,

feature_engine/selection/information_value.py

@@ -9,6 +9,9 @@
     _variables_attribute_docstring,
 )
 from feature_engine._docstrings.init_parameters.encoders import _ignore_format_docstring
+from feature_engine._docstrings.init_parameters.selection import (
+    _confirm_variables_docstring,
+)
 from feature_engine._docstrings.methods import _fit_transform_docstring
 from feature_engine._docstrings.substitute import Substitution
 from feature_engine._variable_handling.init_parameter_checks import (
@@ -43,7 +46,7 @@
     feature_names_in=_feature_names_in_docstring,
     n_features_in=_n_features_in_docstring,
     fit_transform=_fit_transform_docstring,
-    confirm_variables=BaseSelector._confirm_variables_docstring,
+    confirm_variables=_confirm_variables_docstring,
     get_support=_get_support_docstring,
 )
 class SelectByInformationValue(BaseSelector, WoE):