:orphan:

|rss_image|  **onnx - 1/1** :ref:`Blog <ap-main-0>` :ref:`benchmark (4) <ap-cat-benchmark-0>` :ref:`onnx (8) <ap-cat-onnx-0>`

.. |rss_image| image:: feed-icon-16x16.png
    :target: ../_downloads/rss.xml
    :alt: RSS

----


.. index:: onnx


.. _ap-cat-onnx-0:

onnx - 1/1
++++++++++

.. blogpostagg::
    :title: Xop, easy to create onnx graph
    :date: 2022-02-27
    :keywords: tips,tensorflow,tensorflow-onnx
    :categories: xop,onnx
    :rawfile: 2022/2022-02-27_xop.rst

    :epkg:`onnx` package has a very verbose API to create ONNX
    graph. Could you imagine a user to directly write the syntax tree
    of a program instead of some python code? Creating a ONNX graph is
    very similar to that task except ONNX language is more simple
    than python.
    

    ...




.. blogpostagg::
    :title: Decompose einsum into numpy operators
    :date: 2021-08-11
    :keywords: einsum
    :categories: onnx
    :rawfile: 2021/2021-08-11_einsum.rst

    Notebook :ref:`einsumdecompositionrst` what function :epkg:`numpy:einsum`
    does and how it can be decomposed into a series of basic operations,
    all available in ONNX. That's the purpose of function
    Function :func:`decompose_einsum_equation
    <mlprodict.testing.einsum.einsum_impl.decompose_einsum_equation>`.
    With function :func:`export2numpy
    <mlprodict.onnx_tools.onnx_export.export2numpy>`, it is possible to
    convert back this ONNX graph into a series of numpy operations.
    

    ...




.. blogpostagg::
    :title: onnxruntime shape [] != None
    :date: 2021-08-10
    :keywords: onnxruntime
    :categories: onnx
    :rawfile: 2021/2021-08-10_shape.rst

    `None` is the undefined shape, `[]` is an empty shape.
    And when shapes do not fit the results, the outputs can
    be suprising. The following example shows what :epkg:`onnxruntime`
    produces for the same graph except input and output shapes
    when defined as `None` and `[]`.
    

    ...




.. blogpostagg::
    :title: Operator CDist
    :date: 2019-09-16
    :keywords: onnxruntime,cdist
    :categories: onnx
    :rawfile: 2019/2019-09-16_cdist.rst

    Notebooks :ref:`onnxpdistrst` shows how much slower
    an :epkg:`ONNX` implementation of function
    :epkg:`cdist`, from 3 to 10 times slower.
    One way to optimize the converted model is to
    create dedicated operator such as one for function
    :epkg:`cdist`. Tutorial :ref:`l-onnx-tutorial-optim`
    explains how to tell function :func:`to_onnx
    <mlprodict.onnx_conv.convert.to_onnx>` to use
    the custom operator `CDist`.




.. blogpostagg::
    :title: Float, double with ONNX
    :date: 2019-08-23
    :keywords: onnx,float,double
    :categories: onnx
    :rawfile: 2019/2019-08-23_onnx_float_double.rst

    Replicating what a library does, :epkg:`scikit-learn` for
    example, is different from implementing a function
    defined in a paper. Every trick needs to be replicated.
    :epkg:`scikit-learn` trees implement a prediction function
    which takes float features and compares them to double
    thresholds. Knowning the :epkg:`ONNX` assumes that comparison
    only happens numbers of the same type, you end up with discrepencies.
    

    ...




.. blogpostagg::
    :title: ONNX updates
    :date: 2019-08-02
    :keywords: onnx,onnxrt,update
    :categories: onnx
    :rawfile: 2019/2019-08-02_onnx_changes.rst

    The python runtime is now almost complete for
    all the supported numerical operator implemented
    in :epkg:`sklearn-onnx`. A couple of notebooks
    introduces a couple of way to investigates issues,
    to benchmark ONNX models with :epkg:`onnxruntime`
    or python runtime, to check the differences between
    the same model. It also extend ONNX with operators not in
    the specification to experiment some assumptions
    and check it is more efficient. Notebook
    :ref:`onnxshakerrst` introduces a way to guess the
    margins introduced by the conversion from double
    to single. There also exists a function to convert numpy
    function into ONNX (see :ref:`l-numpy2onnx-tutorial`).
    Its coverage is probably low but it will improve.




.. blogpostagg::
    :title: ONNX, runtime
    :date: 2019-06-25
    :keywords: onnx,onnxrt
    :categories: onnx
    :rawfile: 2019/2019-06-25_runtime.rst

    Somebody asked me one day if it would be difficult to
    write a runtime for :epkg:`ONNX` in :epkg:`Rust`.
    I just replied that it should not take that long
    but it would require to implement a way to goes
    through the nodes of the :epkg:`ONNX` graph
    and to have an implementation for every
    :epkg:`ONNX Operators`...
    

    ...




.. blogpostagg::
    :title: ONNX, runtime, converters
    :date: 2019-06-15
    :keywords: onnx,onnxrt
    :categories: onnx
    :rawfile: 2019/2019-06-15_onnxrt.rst

    I have been recently working on :epkg:`sklearn-onnx`
    to write converter from :epkg:`scikit-learn` operators
    to :epkg:`ONNX` serialization format. I was talking
    about that a month ago and somebody asked me if there
    was a runtime implemented in `RUST <https://www.rust-lang.org/>`_.
    Not that I know of but I said it would not be too complex
    to implement one.
    

    ...





----

|rss_image|  **onnx - 1/1** :ref:`2021-08 (3) <ap-month-2021-08-0>` :ref:`2022-02 (1) <ap-month-2022-02-0>` :ref:`2022-05 (1) <ap-month-2022-05-0>` :ref:`2022-06 (1) <ap-month-2022-06-0>` :ref:`2022-11 (1) <ap-month-2022-11-0>`