.. DO NOT EDIT.
.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
.. "auto_examples/plot_mpg_svr.py"
.. LINE NUMBERS ARE GIVEN BELOW.

.. only:: html

    .. note::
        :class: sphx-glr-download-link-note

        :ref:`Go to the end <sphx_glr_download_auto_examples_plot_mpg_svr.py>`
        to download the full example code.

.. rst-class:: sphx-glr-example-title

.. _sphx_glr_auto_examples_plot_mpg_svr.py:


======================================
Plotting Bagging Regression Error Bars
======================================

This example demonstrates using `forestci` to calculate the error bars of
the predictions of a :class:`sklearn.ensemble.BaggingRegressor` object.

The data used here are the Auto MPG dataset by R. Quinlan, bundled under the
Creative Commons Attribution 4.0 International license. See ``data/README.md``
for attribution and licensing details.

.. GENERATED FROM PYTHON SOURCE LINES 13-70



.. image-sg:: /auto_examples/images/sphx_glr_plot_mpg_svr_001.png
   :alt: plot mpg svr
   :srcset: /auto_examples/images/sphx_glr_plot_mpg_svr_001.png
   :class: sphx-glr-single-img





.. code-block:: Python


    # Regression Forest Example
    from pathlib import Path

    import numpy as np
    from matplotlib import pyplot as plt
    from sklearn.ensemble import BaggingRegressor
    from sklearn.svm import SVR
    import sklearn.model_selection as xval
    import forestci as fci

    # Load the bundled Auto MPG data
    data_path = Path.cwd() / "data" / "auto_mpg.csv"
    if not data_path.exists():
        # Also support running ``python examples/plot_mpg_svr.py`` from the repo root.
        data_path = Path.cwd() / "examples" / "data" / "auto_mpg.csv"
    mpg_data = np.genfromtxt(
        data_path,
        delimiter=",",
        skip_header=1,
    )

    # Separate the predictors and target, removing rows with missing values
    mpg_data = mpg_data[~np.isnan(mpg_data).any(axis=1)]
    mpg_X = mpg_data[:, :-1]
    mpg_y = mpg_data[:, -1]

    # Split the data into training and test sets
    X_train, X_test, y_train, y_test = xval.train_test_split(
        mpg_X, mpg_y, test_size=0.25, random_state=42
    )

    # Create a bagged SVR model
    n_estimators = 1000
    bagger = BaggingRegressor(
        estimator=SVR(), n_estimators=n_estimators, random_state=42
    )
    bagger.fit(X_train, y_train)
    y_pred = bagger.predict(X_test)
    target_range = [mpg_y.min(), mpg_y.max()]

    # Plot predictions without error bars
    plt.scatter(y_test, y_pred)
    plt.plot(target_range, target_range, "k--")
    plt.xlabel("Reported MPG")
    plt.ylabel("Predicted MPG")
    plt.show()

    # Calculate the variance
    variance = fci.random_forest_error(bagger, X_train.shape, X_test)

    # Plot error bars for predictions using unbiased variance
    plt.errorbar(y_test, y_pred, yerr=np.sqrt(variance), fmt="o")
    plt.plot(target_range, target_range, "k--")
    plt.xlabel("Reported MPG")
    plt.ylabel("Predicted MPG")
    plt.show()


.. rst-class:: sphx-glr-timing

   **Total running time of the script:** (0 minutes 3.116 seconds)


.. _sphx_glr_download_auto_examples_plot_mpg_svr.py:

.. only:: html

  .. container:: sphx-glr-footer sphx-glr-footer-example

    .. container:: sphx-glr-download sphx-glr-download-jupyter

      :download:`Download Jupyter notebook: plot_mpg_svr.ipynb <plot_mpg_svr.ipynb>`

    .. container:: sphx-glr-download sphx-glr-download-python

      :download:`Download Python source code: plot_mpg_svr.py <plot_mpg_svr.py>`

    .. container:: sphx-glr-download sphx-glr-download-zip

      :download:`Download zipped: plot_mpg_svr.zip <plot_mpg_svr.zip>`


.. only:: html

 .. rst-class:: sphx-glr-signature

    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_