Update pydoc and promote ColumnTransformer out of experimental (#4509)

- Promote `ColumnTransformer` out of experimental - Add `ColumnTransformer` to documentation (answers to #4418) - Update preprocessing documentation Authors: - Victor Lafargue (https://github.com/viclafargue) - Dante Gama Dessavre (https://github.com/dantegd) Approvers: - Micka (https://github.com/lowener) - Dante Gama Dessavre (https://github.com/dantegd) URL: #4509
rapidsai · May 23, 2022 · d9dd8ca · d9dd8ca
1 parent bc43c0e
commit d9dd8ca
Show file tree

Hide file tree

Showing 8 changed files with 90 additions and 83 deletions.
diff --git a/docs/source/api.rst b/docs/source/api.rst
@@ -82,9 +82,46 @@ Feature and Label Encoding (Single-GPU)
  .. autoclass:: cuml.preprocessing.TargetEncoder.TargetEncoder
     :members:
 
+Feature Scaling and Normalization (Single-GPU)
+----------------------------------------------
+.. autoclass:: cuml.preprocessing.MaxAbsScaler
+    :members:
+.. autoclass:: cuml.preprocessing.MinMaxScaler
+    :members:
+.. autoclass:: cuml.preprocessing.Normalizer
+    :members:
+.. autoclass:: cuml.preprocessing.RobustScaler
+    :members:
+.. autoclass:: cuml.preprocessing.StandardScaler
+    :members:
+.. autofunction:: cuml.preprocessing.maxabs_scale
+.. autofunction:: cuml.preprocessing.minmax_scale
+.. autofunction:: cuml.preprocessing.normalize
+.. autofunction:: cuml.preprocessing.robust_scale
+.. autofunction:: cuml.preprocessing.scale
+
+Other preprocessing methods (Single-GPU)
+----------------------------------------
+.. autoclass:: cuml.preprocessing.Binarizer
+    :members:
+.. autoclass:: cuml.preprocessing.FunctionTransformer
+    :members:
+.. autoclass:: cuml.preprocessing.KBinsDiscretizer
+    :members:
+.. autoclass:: cuml.preprocessing.MissingIndicator
+    :members:
+.. autoclass:: cuml.preprocessing.PolynomialFeatures
+    :members:
+.. autoclass:: cuml.preprocessing.SimpleImputer
+    :members:
+.. autofunction:: cuml.preprocessing.add_dummy_feature
+.. autofunction:: cuml.preprocessing.binarize
+
+.. automodule:: cuml.compose
+   :members: ColumnTransformer, make_column_transformer, make_column_selector
 
 Text Preprocessing (Single-GPU)
----------------------------------------
+-------------------------------
  .. autoclass:: cuml.preprocessing.text.stem.PorterStemmer
     :members:
 
@@ -589,14 +626,6 @@ Experimental
     the root `cuml` package. Each `experimental` submodule must be imported
     separately.
 
-Preprocessing
--------------
-.. automodule:: cuml.experimental.preprocessing
-   :members: Binarizer, KBinsDiscretizer, MaxAbsScaler, MinMaxScaler,
-      Normalizer, RobustScaler, SimpleImputer, StandardScaler,
-      add_dummy_feature, binarize, minmax_scale, normalize,
-      PolynomialFeatures, robust_scale, scale
-
 Linear Models
 -------------
 .. autoclass:: cuml.experimental.linear_model.Lars

diff --git a/python/cuml/_thirdparty/sklearn/preprocessing/_column_transformer.py b/python/cuml/_thirdparty/sklearn/preprocessing/_column_transformer.py
@@ -439,7 +439,7 @@ class ColumnTransformer(TransformerMixin, BaseComposition, BaseEstimator):
             its parameters to be set using ``set_params`` and searched in grid
             search.
         transformer : {'drop', 'passthrough'} or estimator
-            Estimator must support :term:`fit` and :term:`transform`.
+            Estimator must support `fit` and `transform`.
             Special-cased strings 'drop' and 'passthrough' are accepted as
             well, to indicate to drop the columns or to pass them through
             untransformed, respectively.
@@ -464,9 +464,9 @@ class ColumnTransformer(TransformerMixin, BaseComposition, BaseEstimator):
         the transformers.
         By setting ``remainder`` to be an estimator, the remaining
         non-specified columns will use the ``remainder`` estimator. The
-        estimator must support :term:`fit` and :term:`transform`.
+        estimator must support `fit` and `transform`.
         Note that using this feature requires that the DataFrame columns
-        input at :term:`fit` and :term:`transform` have identical order.
+        input at `fit` and `transform` have identical order.
 
     sparse_threshold : float, default=0.3
         If the output of the different transformers contains sparse matrices,
@@ -1031,7 +1031,7 @@ def make_column_transformer(*transformers,
         transformer objects to be applied to subsets of the data.
 
         transformer : {'drop', 'passthrough'} or estimator
-            Estimator must support :term:`fit` and :term:`transform`.
+            Estimator must support `fit` and `transform`.
             Special-cased strings 'drop' and 'passthrough' are accepted as
             well, to indicate to drop the columns or to pass them through
             untransformed, respectively.
@@ -1056,7 +1056,7 @@ def make_column_transformer(*transformers,
         the transformers.
         By setting ``remainder`` to be an estimator, the remaining
         non-specified columns will use the ``remainder`` estimator. The
-        estimator must support :term:`fit` and :term:`transform`.
+        estimator must support `fit` and `transform`.
 
     sparse_threshold : float, default=0.3
         If the transformed output consists of a mix of sparse and dense data,
@@ -1069,7 +1069,7 @@ def make_column_transformer(*transformers,
     n_jobs : int, default=None
         Number of jobs to run in parallel.
         ``None`` means 1 unless in a :obj:`joblib.parallel_backend` context.
-        ``-1`` means using all processors. See :term:`Glossary <n_jobs>`
+        ``-1`` means using all processors. See `Glossary <n_jobs>`
         for more details.
 
     verbose : bool, default=False

diff --git a/python/cuml/_thirdparty/sklearn/preprocessing/_discretization.py b/python/cuml/_thirdparty/sklearn/preprocessing/_discretization.py
@@ -93,7 +93,7 @@ class KBinsDiscretizer(TransformerMixin,
       np.concatenate([-np.inf, bin_edges_[i][1:-1], np.inf])
 
     You can combine ``KBinsDiscretizer`` with
-    :class:`sklearn.compose.ColumnTransformer` if you only want to preprocess
+    :class:`cuml.compose.ColumnTransformer` if you only want to preprocess
     part of the features.
 
     ``KBinsDiscretizer`` might produce constant features (e.g., when

diff --git a/python/cuml/compose/__init__.py b/python/cuml/compose/__init__.py
@@ -0,0 +1,27 @@
+#
+# Copyright (c) 2020-2022, NVIDIA CORPORATION.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+
+from cuml._thirdparty.sklearn.preprocessing import ColumnTransformer, \
+    make_column_transformer, make_column_selector
+
+
+__all__ = [
+    # Classes
+    'ColumnTransformer',
+    # Functions
+    'make_column_transformer',
+    'make_column_selector'
+]
diff --git a/python/cuml/experimental/preprocessing/__init__.py b/python/cuml/experimental/preprocessing/__init__.py
diff --git a/python/cuml/preprocessing/__init__.py b/python/cuml/preprocessing/__init__.py
@@ -20,38 +20,42 @@
 from cuml.preprocessing.TargetEncoder import TargetEncoder
 from cuml.preprocessing import text
 
-from cuml._thirdparty.sklearn.preprocessing import StandardScaler, \
-    MinMaxScaler, MaxAbsScaler, Normalizer, Binarizer, PolynomialFeatures, \
-    SimpleImputer, RobustScaler, KBinsDiscretizer, MissingIndicator
-from cuml._thirdparty.sklearn.preprocessing import scale, minmax_scale, \
-    maxabs_scale, normalize, add_dummy_feature, binarize, robust_scale
+from cuml._thirdparty.sklearn.preprocessing import Binarizer, \
+    FunctionTransformer, KBinsDiscretizer, MaxAbsScaler, MinMaxScaler, \
+    MissingIndicator, Normalizer, PolynomialFeatures, RobustScaler, \
+    SimpleImputer, StandardScaler
+
+from cuml._thirdparty.sklearn.preprocessing import add_dummy_feature, \
+    binarize, maxabs_scale, minmax_scale, normalize, robust_scale, scale
+
 
 __all__ = [
     # Classes
     'Binarizer',
+    'FunctionTransformer',
     'KBinsDiscretizer',
+    'LabelBinarizer',
+    'LabelEncoder',
     'MaxAbsScaler',
     'MinMaxScaler',
+    'MissingIndicator',
     'Normalizer',
+    'OneHotEncoder',
     'PolynomialFeatures',
     'RobustScaler',
     'SimpleImputer',
-    'MissingIndicator',
     'StandardScaler',
-    'LabelEncoder',
-    'LabelBinarizer',
-    'OneHotEncoder',
     'TargetEncoder',
     # Functions
     'add_dummy_feature',
     'binarize',
-    'minmax_scale',
+    'label_binarize',
     'maxabs_scale',
+    'minmax_scale',
     'normalize',
     'robust_scale',
     'scale',
-    'label_binarize',
     'train_test_split',
     # Modules
-    'text',
+    'text'
 ]
diff --git a/python/cuml/tests/test_compose.py b/python/cuml/tests/test_compose.py
@@ -20,7 +20,7 @@
 from pandas import DataFrame as pdDataFrame
 from cudf import DataFrame as cuDataFrame
 
-from cuml.experimental.preprocessing import \
+from cuml.compose import \
     ColumnTransformer as cuColumnTransformer, \
     make_column_transformer as cu_make_column_transformer, \
     make_column_selector as cu_make_column_selector

diff --git a/python/cuml/tests/test_preprocessing.py b/python/cuml/tests/test_preprocessing.py
@@ -26,7 +26,7 @@
     RobustScaler as cuRobustScaler, \
     KBinsDiscretizer as cuKBinsDiscretizer, \
     MissingIndicator as cuMissingIndicator
-from cuml.experimental.preprocessing import \
+from cuml.preprocessing import \
     FunctionTransformer as cuFunctionTransformer
 from cuml.preprocessing import scale as cu_scale, \
                     minmax_scale as cu_minmax_scale, \