4.1. Pipelines and composite estimators

Transformers are usually combined with classifiers, regressors or other estimators to build a composite estimator. The most common tool is a Pipeline. Pipeline is often used in combination with FeatureUnion which concatenates the output of transformers into a composite feature space. TransformedTargetRegressor deals with transforming the target (i.e. log-transform y). In contrast, Pipelines only transform the observed data (X).

4.1.1. Pipeline: chaining estimators

Pipeline can be used to chain multiple estimators into one. This is useful as there is often a fixed sequence of steps in processing the data, for example feature selection, normalization and classification. Pipeline serves multiple purposes here:

Convenience and encapsulation
You only have to call fit and predict once on your data to fit a whole sequence of estimators.
Joint parameter selection
You can grid search over parameters of all estimators in the pipeline at once.
Pipelines help avoid leaking statistics from your test data into the trained model in cross-validation, by ensuring that the same samples are used to train the transformers and predictors.

All estimators in a pipeline, except the last one, must be transformers (i.e. must have a transform method). The last estimator may be any type (transformer, classifier, etc.). Usage

The Pipeline is built using a list of (key, value) pairs, where the key is a string containing the name you want to give this step and value is an estimator object:

>>> from sklearn.pipeline import Pipeline
>>> from sklearn.svm import SVC
>>> from sklearn.decomposition import PCA
>>> estimators = [('reduce_dim', PCA()), ('clf', SVC())]
>>> pipe = Pipeline(estimators)
>>> pipe 
         steps=[('reduce_dim', PCA(copy=True,...)),
                ('clf', SVC(C=1.0,...))])

The utility function make_pipeline is a shorthand for constructing pipelines; it takes a variable number of estimators and returns a pipeline, filling in the names automatically:

>>> from sklearn.pipeline import make_pipeline
>>> from sklearn.naive_bayes import MultinomialNB
>>> from sklearn.preprocessing import Binarizer
>>> make_pipeline(Binarizer(), MultinomialNB()) 
         steps=[('binarizer', Binarizer(copy=True, threshold=0.0)),
                ('multinomialnb', MultinomialNB(alpha=1.0,

The estimators of a pipeline are stored as a list in the steps attribute:

>>> pipe.steps[0]
('reduce_dim', PCA(copy=True, iterated_power='auto', n_components=None, random_state=None,
  svd_solver='auto', tol=0.0, whiten=False))

and as a dict in named_steps:

>>> pipe.named_steps['reduce_dim']
PCA(copy=True, iterated_power='auto', n_components=None, random_state=None,
  svd_solver='auto', tol=0.0, whiten=False)

Parameters of the estimators in the pipeline can be accessed using the <estimator>__<parameter> syntax:

>>> pipe.set_params(clf__C=10) 
         steps=[('reduce_dim', PCA(copy=True, iterated_power='auto',...)),
                ('clf', SVC(C=10, cache_size=200, class_weight=None,...))])

Attributes of named_steps map to keys, enabling tab completion in interactive environments:

>>> pipe.named_steps.reduce_dim is pipe.named_steps['reduce_dim']

This is particularly important for doing grid searches:

>>> from sklearn.model_selection import GridSearchCV
>>> param_grid = dict(reduce_dim__n_components=[2, 5, 10],
...                   clf__C=[0.1, 10, 100])
>>> grid_search = GridSearchCV(pipe, param_grid=param_grid)

Individual steps may also be replaced as parameters, and non-final steps may be ignored by setting them to None:

>>> from sklearn.linear_model import LogisticRegression
>>> param_grid = dict(reduce_dim=[None, PCA(5), PCA(10)],
...                   clf=[SVC(), LogisticRegression()],
...                   clf__C=[0.1, 10, 100])
>>> grid_search = GridSearchCV(pipe, param_grid=param_grid) Notes

Calling fit on the pipeline is the same as calling fit on each estimator in turn, transform the input and pass it on to the next step. The pipeline has all the methods that the last estimator in the pipeline has, i.e. if the last estimator is a classifier, the Pipeline can be used as a classifier. If the last estimator is a transformer, again, so is the pipeline. Caching transformers: avoid repeated computation

Fitting transformers may be computationally expensive. With its memory parameter set, Pipeline will cache each transformer after calling fit. This feature is used to avoid computing the fit transformers within a pipeline if the parameters and input data are identical. A typical example is the case of a grid search in which the transformers can be fitted only once and reused for each configuration.

The parameter memory is needed in order to cache the transformers. memory can be either a string containing the directory where to cache the transformers or a joblib.Memory object:

>>> from tempfile import mkdtemp
>>> from shutil import rmtree
>>> from sklearn.decomposition import PCA
>>> from sklearn.svm import SVC
>>> from sklearn.pipeline import Pipeline
>>> estimators = [('reduce_dim', PCA()), ('clf', SVC())]
>>> cachedir = mkdtemp()
>>> pipe = Pipeline(estimators, memory=cachedir)
>>> pipe 
         steps=[('reduce_dim', PCA(copy=True,...)),
                ('clf', SVC(C=1.0,...))])
>>> # Clear the cache directory when you don't need it anymore
>>> rmtree(cachedir)


Side effect of caching transformers

Using a Pipeline without cache enabled, it is possible to inspect the original instance such as:

>>> from sklearn.datasets import load_digits
>>> digits = load_digits()
>>> pca1 = PCA()
>>> svm1 = SVC(gamma='scale')
>>> pipe = Pipeline([('reduce_dim', pca1), ('clf', svm1)])
>>> pipe.fit(digits.data, digits.target)
         steps=[('reduce_dim', PCA(...)), ('clf', SVC(...))])
>>> # The pca instance can be inspected directly
>>> print(pca1.components_) 
    [[-1.77484909e-19  ... 4.07058917e-18]]

Enabling caching triggers a clone of the transformers before fitting. Therefore, the transformer instance given to the pipeline cannot be inspected directly. In following example, accessing the PCA instance pca2 will raise an AttributeError since pca2 will be an unfitted transformer. Instead, use the attribute named_steps to inspect estimators within the pipeline:

>>> cachedir = mkdtemp()
>>> pca2 = PCA()
>>> svm2 = SVC(gamma='scale')
>>> cached_pipe = Pipeline([('reduce_dim', pca2), ('clf', svm2)],
...                        memory=cachedir)
>>> cached_pipe.fit(digits.data, digits.target)
          steps=[('reduce_dim', PCA(...)), ('clf', SVC(...))])
>>> print(cached_pipe.named_steps['reduce_dim'].components_)
    [[-1.77484909e-19  ... 4.07058917e-18]]
>>> # Remove the cache directory
>>> rmtree(cachedir)

4.1.2. Transforming target in regression

TransformedTargetRegressor transforms the targets y before fitting a regression model. The predictions are mapped back to the original space via an inverse transform. It takes as an argument the regressor that will be used for prediction, and the transformer that will be applied to the target variable:

>>> import numpy as np
>>> from sklearn.datasets import load_boston
>>> from sklearn.compose import TransformedTargetRegressor
>>> from sklearn.preprocessing import QuantileTransformer
>>> from sklearn.linear_model import LinearRegression
>>> from sklearn.model_selection import train_test_split
>>> boston = load_boston()
>>> X = boston.data
>>> y = boston.target
>>> transformer = QuantileTransformer(output_distribution='normal')
>>> regressor = LinearRegression()
>>> regr = TransformedTargetRegressor(regressor=regressor,
...                                   transformer=transformer)
>>> X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=0)
>>> regr.fit(X_train, y_train) 
>>> print('R2 score: {0:.2f}'.format(regr.score(X_test, y_test)))
R2 score: 0.67
>>> raw_target_regr = LinearRegression().fit(X_train, y_train)
>>> print('R2 score: {0:.2f}'.format(raw_target_regr.score(X_test, y_test)))
R2 score: 0.64

For simple transformations, instead of a Transformer object, a pair of functions can be passed, defining the transformation and its inverse mapping:

>>> from __future__ import division
>>> def func(x):
...     return np.log(x)
>>> def inverse_func(x):
...     return np.exp(x)

Subsequently, the object is created as:

>>> regr = TransformedTargetRegressor(regressor=regressor,
...                                   func=func,
...                                   inverse_func=inverse_func)
>>> regr.fit(X_train, y_train) 
>>> print('R2 score: {0:.2f}'.format(regr.score(X_test, y_test)))
R2 score: 0.65

By default, the provided functions are checked at each fit to be the inverse of each other. However, it is possible to bypass this checking by setting check_inverse to False:

>>> def inverse_func(x):
...     return x
>>> regr = TransformedTargetRegressor(regressor=regressor,
...                                   func=func,
...                                   inverse_func=inverse_func,
...                                   check_inverse=False)
>>> regr.fit(X_train, y_train) 
>>> print('R2 score: {0:.2f}'.format(regr.score(X_test, y_test)))
R2 score: -4.50


The transformation can be triggered by setting either transformer or the pair of functions func and inverse_func. However, setting both options will raise an error.

4.1.3. FeatureUnion: composite feature spaces

FeatureUnion combines several transformer objects into a new transformer that combines their output. A FeatureUnion takes a list of transformer objects. During fitting, each of these is fit to the data independently. The transformers are applied in parallel, and the feature matrices they output are concatenated side-by-side into a larger matrix.

When you want to apply different transformations to each field of the data, see the related class sklearn.compose.ColumnTransformer (see user guide).

FeatureUnion serves the same purposes as Pipeline - convenience and joint parameter estimation and validation.

FeatureUnion and Pipeline can be combined to create complex models.

(A FeatureUnion has no way of checking whether two transformers might produce identical features. It only produces a union when the feature sets are disjoint, and making sure they are the caller’s responsibility.) Usage

A FeatureUnion is built using a list of (key, value) pairs, where the key is the name you want to give to a given transformation (an arbitrary string; it only serves as an identifier) and value is an estimator object:

>>> from sklearn.pipeline import FeatureUnion
>>> from sklearn.decomposition import PCA
>>> from sklearn.decomposition import KernelPCA
>>> estimators = [('linear_pca', PCA()), ('kernel_pca', KernelPCA())]
>>> combined = FeatureUnion(estimators)
>>> combined 
             transformer_list=[('linear_pca', PCA(copy=True,...)),
                               ('kernel_pca', KernelPCA(alpha=1.0,...))],

Like pipelines, feature unions have a shorthand constructor called make_union that does not require explicit naming of the components.

Like Pipeline, individual steps may be replaced using set_params, and ignored by setting to None:

>>> combined.set_params(kernel_pca=None)
             transformer_list=[('linear_pca', PCA(copy=True,...)),
                               ('kernel_pca', None)],

4.1.4. ColumnTransformer for heterogeneous data


The compose.ColumnTransformer class is experimental and the API is subject to change.

Many datasets contain features of different types, say text, floats, and dates, where each type of feature requires separate preprocessing or feature extraction steps. Often it is easiest to preprocess data before applying scikit-learn methods, for example using pandas. Processing your data before passing it to scikit-learn might be problematic for one of the following reasons:

  1. Incorporating statistics from test data into the preprocessors makes cross-validation scores unreliable (known as data leakage), for example in the case of scalers or imputing missing values.
  2. You may want to include the parameters of the preprocessors in a parameter search.

The ColumnTransformer helps performing different transformations for different columns of the data, within a Pipeline that is safe from data leakage and that can be parametrized. ColumnTransformer works on arrays, sparse matrices, and pandas DataFrames.

To each column, a different transformation can be applied, such as preprocessing or a specific feature extraction method:

>>> import pandas as pd
>>> X = pd.DataFrame(
...     {'city': ['London', 'London', 'Paris', 'Sallisaw'],
...      'title': ["His Last Bow", "How Watson Learned the Trick",
...                "A Moveable Feast", "The Grapes of Wrath"]})

For this data, we might want to encode the 'city' column as a categorical variable, but apply a feature_extraction.text.CountVectorizer to the 'title' column. As we might use multiple feature extraction methods on the same column, we give each transformer a unique name, say 'city_category' and 'title_bow':

>>> from sklearn.compose import ColumnTransformer
>>> from sklearn.feature_extraction.text import CountVectorizer
>>> column_trans = ColumnTransformer(
...     [('city_category', CountVectorizer(analyzer=lambda x: [x]), 'city'),
...      ('title_bow', CountVectorizer(), 'title')])

>>> column_trans.fit(X) 
ColumnTransformer(n_jobs=1, remainder='passthrough', transformer_weights=None,

>>> column_trans.get_feature_names()
['city_category__London', 'city_category__Paris', 'city_category__Sallisaw',
'title_bow__bow', 'title_bow__feast', 'title_bow__grapes', 'title_bow__his',
'title_bow__how', 'title_bow__last', 'title_bow__learned', 'title_bow__moveable',
'title_bow__of', 'title_bow__the', 'title_bow__trick', 'title_bow__watson',

>>> column_trans.transform(X).toarray()
array([[1, 0, 0, 1, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0],
       [1, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 1, 1, 1, 0],
       [0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0],
       [0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 1, 1, 0, 0, 1]]...)

In the above example, the CountVectorizer expects a 1D array as input and therefore the columns were specified as a string ('city'). However, other transformers generally expect 2D data, and in that case you need to specify the column as a list of strings (['city']).

Apart from a scalar or a single item list, the column selection can be specified as a list of multiple items, an integer array, a slice, or a boolean mask. Strings can reference columns if the input is a DataFrame, integers are always interpreted as the positional columns.

The make_columntransformer function is available to more easily create a ColumnTransformer object. Specifically, the names will be given automatically. The equivalent for the above example would be:

>>> from sklearn.compose import make_column_transformer
>>> column_trans = make_column_transformer(
...     ('city', CountVectorizer(analyzer=lambda x: [x])),
...     ('title', CountVectorizer()))
>>> column_trans 
ColumnTransformer(n_jobs=1, remainder='passthrough', transformer_weights=None,
         transformers=[('countvectorizer-1', ...)


  • sphx_glr_auto_examples_compose_column_transformer.py
  • sphx_glr_auto_examples_compose_column_transformer_mixed_types.py