.. _l-pyparis2018:

ONNX: machine learning in production
====================================

.. sharenet::
    :facebook: 1
    :linkedin: 2
    :twitter: 3
    :head: False

.. contents::
    :local:

`PyParis <http://pyparis.org/>`_

Résumé
------

Most of machine learning libraries are optimized to train models and not
necessarily to use them for fast predictions in online web services.
:epkg:`ONNX` is one solution started last year by Microsoft and Facebook.
This presentation describes the concept and shows some examples with
:epkg:`scikit-learn` and :epkg:`ML.net`.

:ref:`onnxdeploypyparisrst`

Links
-----

* :epkg:`ONNX`
* :epkg:`ML.net`
* :epkg:`onnxmltools`
* :epkg:`onnxruntime`
* :epkg:`nimbusml`
* `Custom Extensions to ML.net <http://www.xavierdupre.fr/app/machinelearningext/helpsphinx/index.html>`_

Lectures
--------

* `Optimisation de code avec cffi, numba, cython <http://www.xavierdupre.fr/app/ensae_teaching_cs/helpsphinx/notebooks/cffi_linear_regression.html>`_