.. DO NOT EDIT.
.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
.. "auto_examples_sklearn/plot_sklearn_with_sklearn_onnx.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_sklearn_plot_sklearn_with_sklearn_onnx.py>`
        to download the full example code.

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

.. _sphx_glr_auto_examples_sklearn_plot_sklearn_with_sklearn_onnx.py:


.. _l-plot-sklearn-with-sklearn-onnx:

Using sklearn-onnx to convert any scikit-learn estimator
=========================================================

:func:`yobx.sklearn.to_onnx` ships with built-in converters for a curated
set of :epkg:`scikit-learn` estimators (``StandardScaler``,
``LogisticRegression``, ``DecisionTree*``, ``MLP*``, and ``Pipeline``).
For any estimator that is *not* covered by the built-in registry, you can
supply your own converter through the ``extra_converters`` keyword argument.

:func:`~yobx.sklearn.wrap_skl2onnx_converter` makes the conversion from
:epkg:`sklearn-onnx`.  One line of setup after
looking up the skl2onnx converter function.

.. GENERATED FROM PYTHON SOURCE LINES 17-25

.. code-block:: Python


    import numpy as np
    import onnxruntime
    from sklearn.neural_network import MLPClassifier

    from yobx import doc
    from yobx.sklearn import to_onnx, wrap_skl2onnx_converter








.. GENERATED FROM PYTHON SOURCE LINES 26-30

Converter made for skl2onnx
----------------------------------

We take it from :epkg:`sklearn-onnx`.

.. GENERATED FROM PYTHON SOURCE LINES 30-36

.. code-block:: Python

    import skl2onnx

    convert_sklearn_mlp_classifier = (
        skl2onnx.operator_converters.multilayer_perceptron.convert_sklearn_mlp_classifier
    )








.. GENERATED FROM PYTHON SOURCE LINES 37-42

Convert to ONNX using the factory helper
----------------------------------------

:func:`~yobx.sklearn.wrap_skl2onnx_converter` makes the function look
like a converter for `yobx`.`

.. GENERATED FROM PYTHON SOURCE LINES 42-61

.. code-block:: Python


    X = np.array([[1, 2], [3, 4], [5, 6], [7, 8]], dtype=np.float32)
    y = np.array([0, 0, 1, 1])
    mlp = MLPClassifier(hidden_layer_sizes=(4,), activation="relu", max_iter=2000)
    mlp.fit(X, y)

    converter = wrap_skl2onnx_converter(convert_sklearn_mlp_classifier)
    artifact = to_onnx(mlp, (X,), extra_converters={MLPClassifier: converter})

    ref_b = onnxruntime.InferenceSession(
        artifact.SerializeToString(), providers=["CPUExecutionProvider"]
    )
    results_b = ref_b.run(None, {"X": X})
    label_b, proba_b = results_b[0], results_b[1]

    np.testing.assert_array_equal(mlp.predict(X), label_b)
    np.testing.assert_allclose(mlp.predict_proba(X), proba_b, atol=1e-5)
    print("Predictions match ✓")





.. rst-class:: sphx-glr-script-out

 .. code-block:: none

    Predictions match ✓




.. GENERATED FROM PYTHON SOURCE LINES 62-68

4. Visualise the exported ONNX graph
-------------------------------------

:func:`yobx.doc.plot_dot` renders the :class:`onnx.ModelProto` as a
directed graph using Graphviz.  The nodes emitted by the skl2onnx
converter are injected directly into the enclosing graph.

.. GENERATED FROM PYTHON SOURCE LINES 68-70

.. code-block:: Python


    doc.plot_dot(artifact)



.. image-sg:: /auto_examples_sklearn/images/sphx_glr_plot_sklearn_with_sklearn_onnx_001.png
   :alt: plot sklearn with sklearn onnx
   :srcset: /auto_examples_sklearn/images/sphx_glr_plot_sklearn_with_sklearn_onnx_001.png
   :class: sphx-glr-single-img






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

   **Total running time of the script:** (0 minutes 0.385 seconds)


.. _sphx_glr_download_auto_examples_sklearn_plot_sklearn_with_sklearn_onnx.py:

.. only:: html

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

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

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

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

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

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

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


.. include:: plot_sklearn_with_sklearn_onnx.recommendations


.. only:: html

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

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