Example for preprocessing.dictmapper.DictMapper and `meta.outlier_c…

…lassifier.OutlierClassifier` (koaning#646) * Added an example to preprocessing.dictmapper.DictMapper * Example added to preprocessing.dictmapper.DictMapper * Example added to meta.outlierclassifier.OutlierClassifier * Added example to meta.outlier_classifier.Outlier.Classifier and fixed little mistake I made in outlier_remover * moved examples to class docstring, used compose.ColumnTransformer in dictmapper * Update sklego/preprocessing/dictmapper.py --------- Co-authored-by: Francesco Bruzzesi <[email protected]>
anopsy · Mar 29, 2024 · 14ef241 · 14ef241
1 parent d321198
commit 14ef241
Show file tree

Hide file tree

Showing 4 changed files with 54 additions and 2 deletions.
diff --git a/sklego/meta/outlier_classifier.py b/sklego/meta/outlier_classifier.py
@@ -23,6 +23,30 @@ class OutlierClassifier(BaseEstimator, ClassifierMixin):
         The fitted underlying outlier detection model.
     classes_ : array-like of shape (2,)
         Classes used for prediction (0 or 1)
+
+    Example
+    -------
+    ```py
+    from sklearn.ensemble import IsolationForest
+    from sklego.meta.outlier_classifier import OutlierClassifier
+
+    X = [[0], [0.5], [-1], [99]]
+    y = [0, 0, 0, 1]
+
+    isolation_forest = IsolationForest()
+
+    outlier_clf = OutlierClassifier(isolation_forest)
+    _ = outlier_clf.fit(X, y)
+
+    preds = outlier_clf.predict([[100], [-0.5], [0.5], [1]])
+    # array[1. 0. 0. 0.]
+
+    proba_preds = outlier_clf.predict_proba([[100], [-0.5], [0.5], [1]])
+    # [[0.34946567 0.65053433]
+    #  [0.79707913 0.20292087]
+    #  [0.80275406 0.19724594]
+    #  [0.80275406 0.19724594]]
+    ```
     """
 
     def __init__(self, model):

diff --git a/sklego/model_selection.py b/sklego/model_selection.py
@@ -260,7 +260,7 @@ def KlusterFoldValidation(**kwargs):
 class ClusterFoldValidation:
     """Cross validator that creates folds based on provided cluster method.
     This ensures that data points in the same cluster are not split across different folds.
-    
+
     !!! info "New in version 0.9.0"
 
     Parameters

diff --git a/sklego/preprocessing/dictmapper.py b/sklego/preprocessing/dictmapper.py
@@ -23,6 +23,34 @@ class DictMapper(TransformerMixin, BaseEstimator):
         Number of features seen during `fit`.
     dim_ : int
         Deprecated, please use `n_features_in_` instead.
+
+    Example
+    -------
+    ```py
+    import pandas as pd
+    from sklego.preprocessing.dictmapper import DictMapper
+    from sklearn.compose import ColumnTransformer
+
+    X = pd.DataFrame({
+        "city_pop": ["Amsterdam", "Leiden", "Utrecht", "None", "Haarlem"]
+    })
+
+    mapper = {
+        "Amsterdam": 1_181_817,
+        "Leiden": 130_181,
+        "Utrecht": 367_984,
+        "Haarlem": 165_396,
+    }
+
+    ct = ColumnTransformer([("dictmapper", DictMapper(mapper, 0), ["city_pop"])])
+    X_trans = ct.fit_transform(X)
+    X_trans
+    # array([[1181817],
+    #        [ 130181],
+    #        [ 367984],
+    #        [      0],
+    #        [ 165396]])
+    ```
     """
 
     def __init__(self, mapper, default):

diff --git a/sklego/preprocessing/outlier_remover.py b/sklego/preprocessing/outlier_remover.py
@@ -34,7 +34,7 @@ class OutlierRemover(TrainOnlyTransformerMixin, BaseEstimator):
 
     isolation_forest = IsolationForest()
     isolation_forest.fit(X)
-    detector_preds = isolator_forest.predict(X)
+    detector_preds = isolation_forest.predict(X)
 
     outlier_remover = OutlierRemover(isolation_forest, refit=True)
     outlier_remover.fit(X)