.. _l-sklearn2019:

ONNX: serialization for machine learning models
===============================================

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

`The first workshop of the consortium <https://scikit-learn.fondation-inria.fr/en/may-28-tuesday-the-first-workshop-of-the-consortium/>`_

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 a open format to represent machine learned models.
First developped for deep learning models, it is now available for
:epkg:`scikit-learn` models. It achieves compact representation of models
and fast computation of predictions on CPU, GPU or ARM processors.
The talk introduces three tools to play with ONNX in Python.

* `Slides <https://github.com/xadupre/presentation/blob/master/2019/model_persistence_xavier_dupre.pdf>`_
* :ref:`onnxsklearnconsortiumrst`
* :ref:`onnxsklearncustomrst`

Links
-----

* :epkg:`ONNX`
* :epkg:`sklearn-onnx`
* :epkg:`onnxruntime`