feat: log c plus transformer (#281)

* feat(log): log constant plus transformer

* docs(LogCpTransformer): document LogCpTransformer

* test(test_logcp_transformer): test logcp transformer

* feat(log): add inverse transform method in LogCpTransformer

* test(test_logcp_transformer): inverse transform and C_ parameter

* docs(LogCpTransformer): C_ constant example

* docs(LogCpTransformer): identations

* LogCpTransformer example notebook

* fix(log): C_ as dict for multiple variables

* fix(log): incompatible types in assignment

* method description

* feat(log): log constant plus transformer

* docs(LogCpTransformer): document LogCpTransformer

* test(test_logcp_transformer): test logcp transformer

* feat(log): add inverse transform method in LogCpTransformer

* test(test_logcp_transformer): inverse transform and C_ parameter

* docs(LogCpTransformer): C_ constant example

* docs(LogCpTransformer): identations

* LogCpTransformer example notebook

* fix(log): C_ as dict for multiple variables

* fix(log): incompatible types in assignment

* method description

* rebase master, changes to doscstrings, adds tests for inverse_transform

* modifies base transformer to check variables from dictionary

* modify to pass typechecks

* fixes variable selection from user dictionary passed to c

* merge solegalli branch

* add transformer to readme and main index, minor changes in docstring

Co-authored-by: Soledad Galli <[email protected]>

Author: TremaMiguel

README.md

@@ -89,6 +89,7 @@ More resources will be added as they appear online!
 
 ### Variable Transformation methods
 * LogTransformer
+* LogCpTransformer
 * ReciprocalTransformer
 * PowerTransformer
 * BoxCoxTransformer

docs/images/logcpraw.png

@@ -119,6 +119,7 @@ Numerical Variable Transformation: Transformers
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
 - :doc:`transformation/LogTransformer`: performs logarithmic transformation of numerical variables
+- :doc:`transformation/LogCpTransformer`: adds the variables a constant value and then applies the logarithm
 - :doc:`transformation/ReciprocalTransformer`: performs reciprocal transformation of numerical variables
 - :doc:`transformation/PowerTransformer`: performs power transformation of numerical variables
 - :doc:`transformation/BoxCoxTransformer`: performs Box-Cox transformation of numerical variables

docs/images/logcptransform.png

@@ -0,0 +1,59 @@
+LogCpTransformer
+================
+
+API Reference
+-------------
+
+.. autoclass:: feature_engine.transformation.LogCpTransformer
+    :members:
+
+
+Example
+-------
+
+.. code:: python
+
+    import pandas as pd
+    import matplotlib.pyplot as plt
+    from sklearn.model_selection import train_test_split
+    from sklearn.datasets import load_boston
+
+    from feature_engine import transformation as vt
+
+    # Load dataset
+    X, y = load_boston(return_X_y=True)
+    X = pd.DataFrame(X)
+
+    # Separate into train and test sets
+    X_train, X_test, y_train, y_test =  train_test_split(X, y, test_size=0.3, random_state=0)
+
+    # set up the variable transformer
+    tf = vt.LogCpTransformer(variables = [7, 12], C="auto")
+
+    # fit the transformer
+    tf.fit(X_train)
+
+    # transform the data
+    train_t= tf.transform(X_train)
+    test_t= tf.transform(X_test)
+
+    # learned constant C
+    tf.C_
+
+.. code:: python
+
+    {7: 2.1742, 12: 2.73}
+
+.. code:: python
+
+    # un-transformed variable
+    X_train[12].hist()
+
+.. image:: ../images/logcpraw.png
+
+.. code:: python
+
+    # transformed variable
+    train_t[12].hist()
+
+.. image:: ../images/logcptransform.png

docs/index.rst

@@ -21,7 +21,7 @@ Example
 	from feature_engine import transformation as vt
 
 	# Load dataset
-	data = data = pd.read_csv('houseprice.csv')
+	data = pd.read_csv('houseprice.csv')
 
 	# Separate into train and test sets
 	X_train, X_test, y_train, y_test =  train_test_split(

docs/transformation/LogCpTransformer.rst

@@ -10,6 +10,7 @@ mathematical transformations.
    :maxdepth: 2
 
    LogTransformer
+   LogCpTransformer
    ReciprocalTransformer
    PowerTransformer
    BoxCoxTransformer

docs/transformation/LogTransformer.rst

@@ -2,7 +2,7 @@
 classes. Provides the base functionality within the fit() and transform() methods
 shared by most transformers, like checking that input is a df, the size, NA, etc.
 """
-from typing import List, Optional, Union
+from typing import Dict, Optional
 
 import pandas as pd
 from sklearn.base import BaseEstimator, TransformerMixin
@@ -23,6 +23,47 @@ class BaseNumericalTransformer(BaseEstimator, TransformerMixin):
     variable transformers, discretisers, math combination.
     """
 
+    def _select_variables_from_dict(
+        self, X: pd.DataFrame, user_dict_: Dict
+    ) -> pd.DataFrame:
+        """
+        Checks that input is a dataframe, checks that variables in the dictionary
+        entered by the user are of type numerical.
+
+        Parameters
+        ----------
+        X : Pandas DataFrame
+
+        user_dict_ : Dictionary. Default = None
+            Any dictionary allowed by the transformer and entered by user.
+
+        Raises
+        ------
+        TypeError
+            If the input is not a Pandas DataFrame or a numpy array
+            If any of the variables in the dictionary are not numerical
+        ValueError
+            If there are no numerical variables in the df or the df is empty
+            If the variable(s) contain null values
+
+        Returns
+        -------
+        X : Pandas DataFrame
+            The same dataframe entered as parameter
+        """
+        # check input dataframe
+        X = _is_dataframe(X)
+
+        # find or check for numerical variables
+        variables = [x for x in user_dict_.keys()]
+        self.variables_ = _find_or_check_numerical_variables(X, variables)
+
+        # check if dataset contains na or inf
+        _check_contains_na(X, self.variables_)
+        _check_contains_inf(X, self.variables_)
+
+        return X
+
     def fit(self, X: pd.DataFrame, y: Optional[pd.Series] = None) -> pd.DataFrame:
         """
         Checks that input is a dataframe, finds numerical variables, or alternatively
@@ -54,9 +95,7 @@ def fit(self, X: pd.DataFrame, y: Optional[pd.Series] = None) -> pd.DataFrame:
         X = _is_dataframe(X)
 
         # find or check for numerical variables
-        self.variables_: List[Union[str, int]] = _find_or_check_numerical_variables(
-            X, self.variables
-        )
+        self.variables_ = _find_or_check_numerical_variables(X, self.variables)
 
         # check if dataset contains na or inf
         _check_contains_na(X, self.variables_)

docs/transformation/index.rst

@@ -6,13 +6,7 @@
 import pandas as pd
 
 from feature_engine.base_transformers import BaseNumericalTransformer
-from feature_engine.dataframe_checks import (
-    _check_contains_inf,
-    _check_contains_na,
-    _is_dataframe,
-)
 from feature_engine.validation import _return_tags
-from feature_engine.variable_manipulation import _find_or_check_numerical_variables
 
 
 class ArbitraryDiscretiser(BaseNumericalTransformer):
@@ -131,15 +125,7 @@ def fit(self, X: pd.DataFrame, y: Optional[pd.Series] = None):
         self
         """
         # check input dataframe
-        X = _is_dataframe(X)
-
-        # find or check for numerical variables
-        variables = [x for x in self.binning_dict.keys()]
-        self.variables_ = _find_or_check_numerical_variables(X, variables)
-
-        # check if dataset contains na or inf
-        _check_contains_na(X, self.variables_)
-        _check_contains_inf(X, self.variables_)
+        X = super()._select_variables_from_dict(X, self.binning_dict)
 
         # for consistency wit the rest of the discretisers, we add this attribute
         self.binner_dict_ = self.binning_dict