.. _sphx_glr_auto_examples_plot_cv_predict.py:


====================================
Plotting Cross-Validated Predictions
====================================

This example shows how to use `cross_val_predict` to visualize prediction
errors.




.. image:: /auto_examples/images/sphx_glr_plot_cv_predict_001.png
    :align: center





.. code-block:: python

    from sklearn import datasets
    from sklearn.model_selection import cross_val_predict
    from sklearn import linear_model
    import matplotlib.pyplot as plt

    lr = linear_model.LinearRegression()
    boston = datasets.load_boston()
    y = boston.target

    # cross_val_predict returns an array of the same size as `y` where each entry
    # is a prediction obtained by cross validation:
    predicted = cross_val_predict(lr, boston.data, y, cv=10)

    fig, ax = plt.subplots()
    ax.scatter(y, predicted)
    ax.plot([y.min(), y.max()], [y.min(), y.max()], 'k--', lw=4)
    ax.set_xlabel('Measured')
    ax.set_ylabel('Predicted')
    plt.show()

**Total running time of the script:**
(0 minutes 0.116 seconds)



.. container:: sphx-glr-download

    **Download Python source code:** :download:`plot_cv_predict.py <plot_cv_predict.py>`


.. container:: sphx-glr-download

    **Download IPython notebook:** :download:`plot_cv_predict.ipynb <plot_cv_predict.ipynb>`