.. DO NOT EDIT. .. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY. .. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE: .. "auto_examples/miscellaneous/plot_partial_dependence_visualization_api.py" .. LINE NUMBERS ARE GIVEN BELOW. .. only:: html .. note:: :class: sphx-glr-download-link-note :ref:`Go to the end ` to download the full example code. or to run this example in your browser via Binder .. rst-class:: sphx-glr-example-title .. _sphx_glr_auto_examples_miscellaneous_plot_partial_dependence_visualization_api.py: ========================================= 使用部分依赖的高级绘图 ========================================= :class:`~sklearn.inspection.PartialDependenceDisplay` 对象可以用于绘图，而无需重新计算部分依赖。在这个示例中，我们展示了如何绘制部分依赖图以及如何使用可视化 API 快速自定义图表。 .. NOTE:: 另请参见 :ref:`sphx_glr_auto_examples_miscellaneous_plot_roc_curve_visualization_api.py` .. GENERATED FROM PYTHON SOURCE LINES 12-23 .. code-block:: Python import matplotlib.pyplot as plt import pandas as pd from sklearn.datasets import load_diabetes from sklearn.inspection import PartialDependenceDisplay from sklearn.neural_network import MLPRegressor from sklearn.pipeline import make_pipeline from sklearn.preprocessing import StandardScaler from sklearn.tree import DecisionTreeRegressor .. GENERATED FROM PYTHON SOURCE LINES 24-28 在糖尿病数据集上训练模型 ================================================ 首先，我们在糖尿病数据集上训练一个决策树和一个多层感知器。 .. GENERATED FROM PYTHON SOURCE LINES 28-41 .. code-block:: Python diabetes = load_diabetes() X = pd.DataFrame(diabetes.data, columns=diabetes.feature_names) y = diabetes.target tree = DecisionTreeRegressor() mlp = make_pipeline( StandardScaler(), MLPRegressor(hidden_layer_sizes=(100, 100), tol=1e-2, max_iter=500, random_state=0), ) tree.fit(X, y) mlp.fit(X, y) .. raw:: html

Pipeline(steps=[('standardscaler', StandardScaler()),
                    ('mlpregressor',
                     MLPRegressor(hidden_layer_sizes=(100, 100), max_iter=500,
                                  random_state=0, tol=0.01))])

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.

.. GENERATED FROM PYTHON SOURCE LINES 42-46 绘制两个特征的部分依赖图 ============================ 我们为决策树的特征“age”和“bmi”（身体质量指数）绘制部分依赖曲线。对于两个特征，:func:`~sklearn.inspection.PartialDependenceDisplay.from_estimator` 期望绘制两条曲线。这里，绘图函数使用 `ax` 定义的空间放置了一个包含两个图的网格。 .. GENERATED FROM PYTHON SOURCE LINES 46-50 .. code-block:: Python fig, ax = plt.subplots(figsize=(12, 6)) ax.set_title("Decision Tree") tree_disp = PartialDependenceDisplay.from_estimator(tree, X, ["age", "bmi"], ax=ax) .. image-sg:: /auto_examples/miscellaneous/images/sphx_glr_plot_partial_dependence_visualization_api_001.png :alt: Decision Tree :srcset: /auto_examples/miscellaneous/images/sphx_glr_plot_partial_dependence_visualization_api_001.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 51-52 可以为多层感知机绘制部分依赖曲线。在这种情况下， `line_kw` 被传递给 :func:`~sklearn.inspection.PartialDependenceDisplay.from_estimator` 以更改曲线的颜色。 .. GENERATED FROM PYTHON SOURCE LINES 52-59 .. code-block:: Python fig, ax = plt.subplots(figsize=(12, 6)) ax.set_title("Multi-layer Perceptron") mlp_disp = PartialDependenceDisplay.from_estimator( mlp, X, ["age", "bmi"], ax=ax, line_kw={"color": "red"} ) .. image-sg:: /auto_examples/miscellaneous/images/sphx_glr_plot_partial_dependence_visualization_api_002.png :alt: Multi-layer Perceptron :srcset: /auto_examples/miscellaneous/images/sphx_glr_plot_partial_dependence_visualization_api_002.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 60-67 绘制两个模型的部分依赖图 ========================== `tree_disp` 和 `mlp_disp` :class:`~sklearn.inspection.PartialDependenceDisplay` 对象包含了重新创建部分依赖曲线所需的所有计算信息。这意味着我们可以轻松创建额外的图表，而无需重新计算曲线。一种绘制曲线的方法是将它们放在同一个图中，每个模型的曲线位于每一行。首先，我们创建一个包含两行一列的图形，并在其中放置两个坐标轴。这两个坐标轴将传递给 `tree_disp` 和 `mlp_disp` 的 :func:`~sklearn.inspection.PartialDependenceDisplay.plot` 函数。给定的坐标轴将被绘图函数用来绘制部分依赖图。最终的图将决策树的部分依赖曲线放在第一行，多层感知器的部分依赖曲线放在第二行。 .. GENERATED FROM PYTHON SOURCE LINES 67-74 .. code-block:: Python fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(10, 10)) tree_disp.plot(ax=ax1) ax1.set_title("Decision Tree") mlp_disp.plot(ax=ax2, line_kw={"color": "red"}) ax2.set_title("Multi-layer Perceptron") .. image-sg:: /auto_examples/miscellaneous/images/sphx_glr_plot_partial_dependence_visualization_api_003.png :alt: Decision Tree, Multi-layer Perceptron :srcset: /auto_examples/miscellaneous/images/sphx_glr_plot_partial_dependence_visualization_api_003.png :class: sphx-glr-single-img .. rst-class:: sphx-glr-script-out .. code-block:: none Text(0.5, 1.0, 'Multi-layer Perceptron') .. GENERATED FROM PYTHON SOURCE LINES 75-76 另一种比较曲线的方法是将它们叠加在一起绘制。在这里，我们创建一个包含一行两列的图形。将坐标轴作为列表传递给 :func:`~sklearn.inspection.PartialDependenceDisplay.plot` 函数，该函数将在相同的坐标轴上绘制每个模型的部分依赖曲线。坐标轴列表的长度必须等于绘制的图的数量。 .. GENERATED FROM PYTHON SOURCE LINES 76-86 .. code-block:: Python fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(10, 6)) tree_disp.plot(ax=[ax1, ax2], line_kw={"label": "Decision Tree"}) mlp_disp.plot( ax=[ax1, ax2], line_kw={"label": "Multi-layer Perceptron", "color": "red"} ) ax1.legend() ax2.legend() .. image-sg:: /auto_examples/miscellaneous/images/sphx_glr_plot_partial_dependence_visualization_api_004.png :alt: plot partial dependence visualization api :srcset: /auto_examples/miscellaneous/images/sphx_glr_plot_partial_dependence_visualization_api_004.png :class: sphx-glr-single-img .. rst-class:: sphx-glr-script-out .. code-block:: none .. GENERATED FROM PYTHON SOURCE LINES 88-89 `tree_disp.axes_` 是一个 numpy 数组，包含用于绘制部分依赖图的坐标轴。可以将其传递给 `mlp_disp` ，以实现将图形叠加绘制的效果。此外， `mlp_disp.figure_` 存储了图形对象，这允许在调用 `plot` 之后调整图形的大小。在这种情况下， `tree_disp.axes_` 有两个维度，因此 `plot` 只会在最左边的图上显示 y 轴标签和 y 轴刻度。 .. GENERATED FROM PYTHON SOURCE LINES 89-100 .. code-block:: Python tree_disp.plot(line_kw={"label": "Decision Tree"}) mlp_disp.plot( line_kw={"label": "Multi-layer Perceptron", "color": "red"}, ax=tree_disp.axes_ ) tree_disp.figure_.set_size_inches(10, 6) tree_disp.axes_[0, 0].legend() tree_disp.axes_[0, 1].legend() plt.show() .. image-sg:: /auto_examples/miscellaneous/images/sphx_glr_plot_partial_dependence_visualization_api_005.png :alt: plot partial dependence visualization api :srcset: /auto_examples/miscellaneous/images/sphx_glr_plot_partial_dependence_visualization_api_005.png :class: sphx-glr-single-img .. GENERATED FROM PYTHON SOURCE LINES 101-105 绘制单个特征的部分依赖图 =========================== 在这里，我们在同一坐标轴上绘制单个特征“年龄”的部分依赖曲线。在这种情况下， `tree_disp.axes_` 被传递到第二个绘图函数中。 .. GENERATED FROM PYTHON SOURCE LINES 105-109 .. code-block:: Python tree_disp = PartialDependenceDisplay.from_estimator(tree, X, ["age"]) mlp_disp = PartialDependenceDisplay.from_estimator( mlp, X, ["age"], ax=tree_disp.axes_, line_kw={"color": "red"} ) .. image-sg:: /auto_examples/miscellaneous/images/sphx_glr_plot_partial_dependence_visualization_api_006.png :alt: plot partial dependence visualization api :srcset: /auto_examples/miscellaneous/images/sphx_glr_plot_partial_dependence_visualization_api_006.png :class: sphx-glr-single-img .. rst-class:: sphx-glr-timing **Total running time of the script:** (0 minutes 1.289 seconds) .. _sphx_glr_download_auto_examples_miscellaneous_plot_partial_dependence_visualization_api.py: .. only:: html .. container:: sphx-glr-footer sphx-glr-footer-example .. container:: binder-badge .. image:: images/binder_badge_logo.svg :target: https://mybinder.org/v2/gh/scikit-learn/scikit-learn/main?urlpath=lab/tree/notebooks/auto_examples/miscellaneous/plot_partial_dependence_visualization_api.ipynb :alt: Launch binder :width: 150 px .. container:: sphx-glr-download sphx-glr-download-jupyter :download:`Download Jupyter notebook: plot_partial_dependence_visualization_api.ipynb ` .. container:: sphx-glr-download sphx-glr-download-python :download:`Download Python source code: plot_partial_dependence_visualization_api.py ` .. container:: sphx-glr-download sphx-glr-download-zip :download:`Download zipped: plot_partial_dependence_visualization_api.zip ` .. include:: plot_partial_dependence_visualization_api.recommendations .. only:: html .. rst-class:: sphx-glr-signature `Gallery generated by Sphinx-Gallery `_