.. _imscikitplotrst:

===========
scikit-plot
===========


.. only:: html

    **Links:** :download:`notebook <im_scikit_plot.ipynb>`, :downloadlink:`html <im_scikit_plot2html.html>`, :download:`PDF <im_scikit_plot.pdf>`, :download:`python <im_scikit_plot.py>`, :downloadlink:`slides <im_scikit_plot.slides.html>`, :githublink:`GitHub|_doc/notebooks/2016/pydata/im_scikit_plot.ipynb|*`


*scikit-plot* is an extension of `matplotlib <http://matplotlib.org/>`__
for datascientist. Proposed graphs are a frequent need when playing with
data.

`documentation <https://scikit-plot.readthedocs.io/en/stable/>`__
`source <https://github.com/reiinakano/scikit-plot>`__
`installation <https://scikit-plot.readthedocs.io/en/stable/Quickstart.html?highlight=installation>`__
`tutorial <https://scikit-plot.readthedocs.io/en/stable/Quickstart.html?highlight=installation#your-first-plot>`__

.. code:: ipython3

    %matplotlib inline

.. code:: ipython3

    from sklearn.datasets import load_iris
    data= load_iris()
    X, y = data.data, data.target

.. code:: ipython3

    from sklearn.model_selection import train_test_split
    X_train, X_test, y_train, y_test = train_test_split(X, y)

.. code:: ipython3

    import scikitplot as skplt
    from sklearn.linear_model import LogisticRegression
    nb = LogisticRegression(solver='lbfgs', multi_class='ovr')
    nb = nb.fit(X_train, y_train)
    y_probas = nb.predict_proba(X_test)
    skplt.metrics.plot_roc(y_test, y_probas);



.. image:: im_scikit_plot_5_0.png