Note

Go to the end to download the full example code. or to run this example in your browser via JupyterLite or Binder

Displaying Pipelines#

The default configuration for displaying a pipeline in a Jupyter Notebook is 'diagram' where set_config(display='diagram'). To deactivate HTML representation, use set_config(display='text').

To see more detailed steps in the visualization of the pipeline, click on the steps in the pipeline.

# Authors: The scikit-learn developers
# SPDX-License-Identifier: BSD-3-Clause

Displaying a Pipeline with a Preprocessing Step and Classifier#

This section constructs a Pipeline with a preprocessing step, StandardScaler, and classifier, LogisticRegression, and displays its visual representation.

fromsklearnimport set_config
fromsklearn.linear_modelimport LogisticRegression
fromsklearn.pipelineimport Pipeline
fromsklearn.preprocessingimport StandardScaler
steps = [
 ("preprocessing", StandardScaler ()),
 ("classifier", LogisticRegression ()),
]
pipe = Pipeline (steps)

To visualize the diagram, the default is display='diagram'.

set_config (display="diagram")
pipe # click on the diagram below to see the details of each step

Pipeline(steps=[('preprocessing', StandardScaler()),
 ('classifier', LogisticRegression())])
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

AltStyle によって変換されたページ (->オリジナル) /



To view the text pipeline, change to display='text'.

set_config (display="text")
pipe

Pipeline(steps=[('preprocessing', StandardScaler()),
 ('classifier', LogisticRegression())])

Put back the default display

set_config (display="diagram")

Displaying a Pipeline Chaining Multiple Preprocessing Steps & Classifier#

This section constructs a Pipeline with multiple preprocessing steps, PolynomialFeatures and StandardScaler, and a classifier step, LogisticRegression, and displays its visual representation.

fromsklearn.linear_modelimport LogisticRegression
fromsklearn.pipelineimport Pipeline
fromsklearn.preprocessingimport PolynomialFeatures , StandardScaler
steps = [
 ("standard_scaler", StandardScaler ()),
 ("polynomial", PolynomialFeatures (degree=3)),
 ("classifier", LogisticRegression (C=2.0)),
]
pipe = Pipeline (steps)
pipe # click on the diagram below to see the details of each step

Pipeline(steps=[('standard_scaler', StandardScaler()),
 ('polynomial', PolynomialFeatures(degree=3)),
 ('classifier', LogisticRegression(C=2.0))])
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.


Displaying a Pipeline and Dimensionality Reduction and Classifier#

This section constructs a Pipeline with a dimensionality reduction step, PCA, a classifier, SVC, and displays its visual representation.

fromsklearn.decompositionimport PCA
fromsklearn.pipelineimport Pipeline
fromsklearn.svmimport SVC
steps = [("reduce_dim", PCA (n_components=4)), ("classifier", SVC (kernel="linear"))]
pipe = Pipeline (steps)
pipe # click on the diagram below to see the details of each step

Pipeline(steps=[('reduce_dim', PCA(n_components=4)),
 ('classifier', SVC(kernel='linear'))])
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.


Displaying a Complex Pipeline Chaining a Column Transformer#

This section constructs a complex Pipeline with a ColumnTransformer and a classifier, LogisticRegression, and displays its visual representation.

importnumpyasnp
fromsklearn.composeimport ColumnTransformer
fromsklearn.imputeimport SimpleImputer
fromsklearn.linear_modelimport LogisticRegression
fromsklearn.pipelineimport Pipeline , make_pipeline
fromsklearn.preprocessingimport OneHotEncoder , StandardScaler
numeric_preprocessor = Pipeline (
 steps=[
 ("imputation_mean", SimpleImputer (missing_values=np.nan , strategy="mean")),
 ("scaler", StandardScaler ()),
 ]
)
categorical_preprocessor = Pipeline (
 steps=[
 (
 "imputation_constant",
 SimpleImputer (fill_value="missing", strategy="constant"),
 ),
 ("onehot", OneHotEncoder (handle_unknown="ignore")),
 ]
)
preprocessor = ColumnTransformer (
 [
 ("categorical", categorical_preprocessor, ["state", "gender"]),
 ("numerical", numeric_preprocessor, ["age", "weight"]),
 ]
)
pipe = make_pipeline (preprocessor, LogisticRegression (max_iter=500))
pipe # click on the diagram below to see the details of each step

Pipeline(steps=[('columntransformer',
 ColumnTransformer(transformers=[('categorical',
 Pipeline(steps=[('imputation_constant',
 SimpleImputer(fill_value='missing',
 strategy='constant')),
 ('onehot',
 OneHotEncoder(handle_unknown='ignore'))]),
 ['state', 'gender']),
 ('numerical',
 Pipeline(steps=[('imputation_mean',
 SimpleImputer()),
 ('scaler',
 StandardScaler())]),
 ['age', 'weight'])])),
 ('logisticregression', LogisticRegression(max_iter=500))])
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.


Displaying a Grid Search over a Pipeline with a Classifier#

This section constructs a GridSearchCV over a Pipeline with RandomForestClassifier and displays its visual representation.

importnumpyasnp
fromsklearn.composeimport ColumnTransformer
fromsklearn.ensembleimport RandomForestClassifier
fromsklearn.imputeimport SimpleImputer
fromsklearn.model_selectionimport GridSearchCV
fromsklearn.pipelineimport Pipeline , make_pipeline
fromsklearn.preprocessingimport OneHotEncoder , StandardScaler
numeric_preprocessor = Pipeline (
 steps=[
 ("imputation_mean", SimpleImputer (missing_values=np.nan , strategy="mean")),
 ("scaler", StandardScaler ()),
 ]
)
categorical_preprocessor = Pipeline (
 steps=[
 (
 "imputation_constant",
 SimpleImputer (fill_value="missing", strategy="constant"),
 ),
 ("onehot", OneHotEncoder (handle_unknown="ignore")),
 ]
)
preprocessor = ColumnTransformer (
 [
 ("categorical", categorical_preprocessor, ["state", "gender"]),
 ("numerical", numeric_preprocessor, ["age", "weight"]),
 ]
)
pipe = Pipeline (
 steps=[("preprocessor", preprocessor), ("classifier", RandomForestClassifier ())]
)
param_grid = {
 "classifier__n_estimators": [200, 500],
 "classifier__max_features": ["auto", "sqrt", "log2"],
 "classifier__max_depth": [4, 5, 6, 7, 8],
 "classifier__criterion": ["gini", "entropy"],
}
grid_search = GridSearchCV (pipe, param_grid=param_grid, n_jobs=1)
grid_search # click on the diagram below to see the details of each step

GridSearchCV(estimator=Pipeline(steps=[('preprocessor',
 ColumnTransformer(transformers=[('categorical',
 Pipeline(steps=[('imputation_constant',
 SimpleImputer(fill_value='missing',
 strategy='constant')),
 ('onehot',
 OneHotEncoder(handle_unknown='ignore'))]),
 ['state',
 'gender']),
 ('numerical',
 Pipeline(steps=[('imputation_mean',
 SimpleImputer()),
 ('scaler',
 StandardScaler())]),
 ['age',
 'weight'])])),
 ('classifier',
 RandomForestClassifier())]),
 n_jobs=1,
 param_grid={'classifier__criterion': ['gini', 'entropy'],
 'classifier__max_depth': [4, 5, 6, 7, 8],
 'classifier__max_features': ['auto', 'sqrt', 'log2'],
 'classifier__n_estimators': [200, 500]})
In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.


Total running time of the script: (0 minutes 0.127 seconds)

Related examples

Displaying estimators and complex pipelines

Displaying estimators and complex pipelines

Column Transformer with Mixed Types

Column Transformer with Mixed Types

Introducing the set_output API

Introducing the set_output API

Comparing Target Encoder with Other Encoders

Comparing Target Encoder with Other Encoders

Gallery generated by Sphinx-Gallery