:orphan:

.. _examples-gallery:

Examples Gallery
================





.. raw:: html

    <div class="sphx-glr-thumbnails">


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="ONNX operators does not contain operator from numpy. There is no operator for `solve &lt;https://n...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_qextend_onnxruntime_thumb.png
    :alt: Fast runtime with onnxruntime

  :ref:`sphx_glr_gyexamples_plot_qextend_onnxruntime.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Fast runtime with onnxruntime</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This short code compares the execution of a couple of runtime for inference including onnxrunti...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_benchmark_inference_standard_thumb.png
    :alt: Benchmark inference for scikit-learn models

  :ref:`sphx_glr_gyexamples_plot_benchmark_inference_standard.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Benchmark inference for scikit-learn models</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Every library is versioned. scikit-learn may change the implementation of a specific model. Tha...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_cbegin_opset_thumb.png
    :alt: What is the opset number?

  :ref:`sphx_glr_gyexamples_plot_cbegin_opset.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">What is the opset number?</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This program starts from an example in scikit-learn documentation: `Plot individual and voting ...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_abegin_convert_pipeline_thumb.png
    :alt: Train and deploy a scikit-learn pipeline

  :ref:`sphx_glr_gyexamples_plot_abegin_convert_pipeline.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Train and deploy a scikit-learn pipeline</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="sklearn-onnx only converts scikit-learn models into *ONNX* but many libraries implement scikit-...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_gexternal_lightgbm_thumb.png
    :alt: Convert a pipeline with a LightGBM classifier

  :ref:`sphx_glr_gyexamples_plot_gexternal_lightgbm.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Convert a pipeline with a LightGBM classifier</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Once a model is converted it can be useful to store an array as a constant in the graph an retr...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_gbegin_cst_thumb.png
    :alt: Store arrays in one onnx graph

  :ref:`sphx_glr_gyexamples_plot_gbegin_cst.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Store arrays in one onnx graph</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="There are many reasons why a user wants more than using the converted model into ONNX. Intermed...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_fbegin_investigate_thumb.png
    :alt: Intermediate results and investigation

  :ref:`sphx_glr_gyexamples_plot_fbegin_investigate.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Intermediate results and investigation</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Some runtimes do not implement a runtime for every available operator in ONNX. The converter do...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_dbegin_options_list_thumb.png
    :alt: Black list operators when converting

  :ref:`sphx_glr_gyexamples_plot_dbegin_options_list.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Black list operators when converting</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This example leverages example l-orttraining-linreg-gpu to train a neural network from scikit-l...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_orttraining_nn_gpu_fwbw_thumb.png
    :alt: Forward backward on a neural network on GPU

  :ref:`sphx_glr_gyexamples_plot_orttraining_nn_gpu_fwbw.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Forward backward on a neural network on GPU</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This example leverages example l-orttraining-linreg-gpu to train a neural network from scikit-l...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_orttraining_nn_gpu_thumb.png
    :alt: Train a scikit-learn neural network with onnxruntime-training on GPU

  :ref:`sphx_glr_gyexamples_plot_orttraining_nn_gpu.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Train a scikit-learn neural network with onnxruntime-training on GPU</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="WOE means Weights of Evidence. It consists in checking that a feature X belongs to a series of ...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_woe_transformer_thumb.png
    :alt: Converter for WOE

  :ref:`sphx_glr_gyexamples_plot_woe_transformer.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Converter for WOE</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This example shows how to change the default ONNX graph such as renaming the inputs or outputs ...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_gconverting_thumb.png
    :alt: Modify the ONNX graph

  :ref:`sphx_glr_gyexamples_plot_gconverting.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Modify the ONNX graph</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This example rewrites l-orttraining-linreg with another optimizer onnxcustom.training.optimizer...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_orttraining_linear_regression_fwbw_thumb.png
    :alt: Train a linear regression with forward backward

  :ref:`sphx_glr_gyexamples_plot_orttraining_linear_regression_fwbw.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Train a linear regression with forward backward</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Example l-simple-deploy-1 converts a simple model. This example takes a similar example but on ...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_bbegin_measure_time_thumb.png
    :alt: Benchmark ONNX conversion

  :ref:`sphx_glr_gyexamples_plot_bbegin_measure_time.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Benchmark ONNX conversion</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="A pipeline usually ingests data as a matrix. It may be converted in a matrix if all the data sh...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_gbegin_dataframe_thumb.png
    :alt: Dataframe as an input

  :ref:`sphx_glr_gyexamples_plot_gbegin_dataframe.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Dataframe as an input</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Function sigmoid is sig(x) = \frac{1}{1 + e^{-x}}. For small or high value, implementation has ...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_funny_sigmoid_thumb.png
    :alt: Funny discrepancies

  :ref:`sphx_glr_gyexamples_plot_funny_sigmoid.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Funny discrepancies</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This example explores how onnxruntime-training can be used to train a simple linear regression ...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_orttraining_linear_regression_thumb.png
    :alt: Train a linear regression with onnxruntime-training

  :ref:`sphx_glr_gyexamples_plot_orttraining_linear_regression.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Train a linear regression with onnxruntime-training</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip=" .. contents::     :local:">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_benchmark_onnx_serialize_thumb.png
    :alt: SerializeToString and ParseFromString

  :ref:`sphx_glr_gyexamples_plot_benchmark_onnx_serialize.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">SerializeToString and ParseFromString</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This example does the same as l-orttraining-nn-gpu-fwbw but updates the weights using `Nesterov...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_orttraining_nn_gpu_fwbw_nesterov_thumb.png
    :alt: Forward backward on a neural network on GPU (Nesterov) and penalty

  :ref:`sphx_glr_gyexamples_plot_orttraining_nn_gpu_fwbw_nesterov.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Forward backward on a neural network on GPU (Nesterov) and penalty</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="In many cases, a custom models leverages existing models which already have an associated conve...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_kcustom_converter_wrapper_thumb.png
    :alt: Implement a new converter using other converters

  :ref:`sphx_glr_gyexamples_plot_kcustom_converter_wrapper.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Implement a new converter using other converters</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="The goal is to compare the processing time between batch predictions and one-off prediction for...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_gexternal_lightgbm_reg_per_thumb.png
    :alt: Batch predictions vs one-off predictions

  :ref:`sphx_glr_gyexamples_plot_gexternal_lightgbm_reg_per.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Batch predictions vs one-off predictions</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="By default, sklearn-onnx assumes that a classifier has two outputs (label and probabilities), a...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_mcustom_parser_thumb.png
    :alt: Change the number of outputs by adding a parser

  :ref:`sphx_glr_gyexamples_plot_mcustom_parser.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Change the number of outputs by adding a parser</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="The following examples converts a model into ONNX and runs it with onnxruntime. This one is the...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_profile_ort_thumb.png
    :alt: Profile onnxruntime execution

  :ref:`sphx_glr_gyexamples_plot_profile_ort.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Profile onnxruntime execution</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="There is not one way to convert a model. A new operator might have been added in a newer versio...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_dbegin_options_thumb.png
    :alt: One model, many possible conversions with options

  :ref:`sphx_glr_gyexamples_plot_dbegin_options.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">One model, many possible conversions with options</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="There are two ways to write a converter. The first one is very verbose (see `ada_boost.py &lt;http...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_jcustom_syntax_thumb.png
    :alt: Two ways to implement a converter

  :ref:`sphx_glr_gyexamples_plot_jcustom_syntax.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Two ways to implement a converter</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="A scikit-learn classifier usually returns a matrix of probabilities. By default, *sklearn-onnx*...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_dbegin_options_zipmap_thumb.png
    :alt: Choose appropriate output of a classifier

  :ref:`sphx_glr_gyexamples_plot_dbegin_options_zipmap.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Choose appropriate output of a classifier</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Options are used to implement different conversion for a same model. The options can be used to...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_lcustom_options_thumb.png
    :alt: A new converter with options

  :ref:`sphx_glr_gyexamples_plot_lcustom_options.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">A new converter with options</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Transfer learning is common with deep learning. A deep learning model is used as preprocessing ...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_gbegin_transfer_learning_thumb.png
    :alt: Transfer Learning with ONNX

  :ref:`sphx_glr_gyexamples_plot_gbegin_transfer_learning.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Transfer Learning with ONNX</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="sklearn-onnx only converts scikit-learn models into ONNX but many libraries implement scikit-le...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_gexternal_xgboost_thumb.png
    :alt: Convert a pipeline with a XGBoost model

  :ref:`sphx_glr_gyexamples_plot_gexternal_xgboost.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Convert a pipeline with a XGBoost model</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This example shows to profile the execution of an ONNX file with onnxruntime to find the operat...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_profile_ort_onnx_thumb.png
    :alt: Profiling of ONNX graph with onnxruntime

  :ref:`sphx_glr_gyexamples_plot_profile_ort_onnx.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Profiling of ONNX graph with onnxruntime</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="The goal is to compare the processing time for batch predictions with onnxruntime and lightgbm ...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_gexternal_lightgbm_reg_mono_thumb.png
    :alt: Batch predictions without parallelization

  :ref:`sphx_glr_gyexamples_plot_gexternal_lightgbm_reg_mono.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Batch predictions without parallelization</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="The discrepancies observed when using float and TreeEnsemble operator (see l-example-discrepenc...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_gexternal_lightgbm_reg_thumb.png
    :alt: Convert a pipeline with a LightGBM regressor

  :ref:`sphx_glr_gyexamples_plot_gexternal_lightgbm_reg.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Convert a pipeline with a LightGBM regressor</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="By default, sklearn-onnx assumes that a classifier has two outputs (label and probabilities), a...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_icustom_converter_thumb.png
    :alt: Implement a new converter

  :ref:`sphx_glr_gyexamples_plot_icustom_converter.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Implement a new converter</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Most models in scikit-learn do computation with double, not float. Most models in deep learning...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_ebegin_float_double_thumb.png
    :alt: Issues when switching to float

  :ref:`sphx_glr_gyexamples_plot_ebegin_float_double.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Issues when switching to float</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="onnxruntime does optimize the ONNX graph before running the inference. It tries for example to ...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_benchmark_graph_opt_thumb.png
    :alt: Benchmark onnxruntime optimization

  :ref:`sphx_glr_gyexamples_plot_benchmark_graph_opt.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Benchmark onnxruntime optimization</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="The benchmark compares the processing time between scikit-learn and onnxruntime-training on a l...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_orttraining_benchmark_fwbw_thumb.png
    :alt: Benchmark, comparison scikit-learn - forward-backward

  :ref:`sphx_glr_gyexamples_plot_orttraining_benchmark_fwbw.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Benchmark, comparison scikit-learn - forward-backward</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="The benchmark compares the processing time between scikit-learn and onnxruntime-training on a l...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_orttraining_benchmark_thumb.png
    :alt: Benchmark, comparison scikit-learn - onnxruntime-training

  :ref:`sphx_glr_gyexamples_plot_orttraining_benchmark.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Benchmark, comparison scikit-learn - onnxruntime-training</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="The operator LeakyRelu is equivalent to the function: LeayRelu(x) = \begin{array}{l} x \text{ i...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_benchmark_op_leakyrelu_thumb.png
    :alt: Benchmark operator LeakyRelu

  :ref:`sphx_glr_gyexamples_plot_benchmark_op_leakyrelu.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Benchmark operator LeakyRelu</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="`TfidfVectorizer &lt;https://scikit-learn.org/stable/modules/ generated/sklearn.feature_extraction...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_transformer_discrepancy_thumb.png
    :alt: Dealing with discrepancies (tf-idf)

  :ref:`sphx_glr_gyexamples_plot_transformer_discrepancy.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Dealing with discrepancies (tf-idf)</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This short code compares the execution of the operator *Slice* on CPU and GPU in three configur...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_benchmark_op_short_thumb.png
    :alt: Benchmark operator Slice

  :ref:`sphx_glr_gyexamples_plot_benchmark_op_short.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Benchmark operator Slice</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="`WOEEncoder &lt;https://contrib.scikit-learn.org/category_encoders/woe.html&gt;`_ is a transformer im...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_catwoe_transformer_thumb.png
    :alt: Converter for WOEEncoder from categorical_encoder

  :ref:`sphx_glr_gyexamples_plot_catwoe_transformer.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Converter for WOEEncoder from categorical_encoder</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Quantization aims at reducing the model size but it does compute the output at a lower precisio...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_quantization_thumb.png
    :alt: Quantization with onnxruntime

  :ref:`sphx_glr_gyexamples_plot_quantization.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Quantization with onnxruntime</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This short code compares different methods to call onnxruntime API.">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_benchmark_ort_api_thumb.png
    :alt: Benchmark onnxruntime API: run or run_with_ort_values

  :ref:`sphx_glr_gyexamples_plot_benchmark_ort_api.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Benchmark onnxruntime API: run or run_with_ort_values</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="The benchmark compares the processing time between pytorch and onnxruntime-training on a linear...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_orttraining_benchmark_torch_thumb.png
    :alt: Benchmark, comparison torch - forward-backward

  :ref:`sphx_glr_gyexamples_plot_orttraining_benchmark_torch.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Benchmark, comparison torch - forward-backward</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="The benchmark compares the processing time between scikit-learn and onnxruntime-training on a l...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_orttraining_benchmark_fwbw_cls_thumb.png
    :alt: Benchmark, comparison sklearn - forward-backward - classification

  :ref:`sphx_glr_gyexamples_plot_orttraining_benchmark_fwbw_cls.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Benchmark, comparison sklearn - forward-backward - classification</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="`TfidfVectorizer &lt;https://scikit-learn.org/stable/modules/ generated/sklearn.feature_extraction...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_usparse_xgboost_thumb.png
    :alt: TfIdf and sparse matrices

  :ref:`sphx_glr_gyexamples_plot_usparse_xgboost.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">TfIdf and sparse matrices</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This short code compares the execution of a couple of runtime for inference including onnxrunti...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_benchmark_inference_thumb.png
    :alt: Benchmark inference for a linear regression

  :ref:`sphx_glr_gyexamples_plot_benchmark_inference.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Benchmark inference for a linear regression</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="ONNX operators do not contain all operators from numpy. There is no operator for `solve &lt;https:...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_pextend_python_runtime_thumb.png
    :alt: Fast design with a python runtime

  :ref:`sphx_glr_gyexamples_plot_pextend_python_runtime.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Fast design with a python runtime</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="onnxruntime can be used a replacement to numpy. It can be used to implement a training algorith...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_benchmark_onnx_function_thumb.png
    :alt: Compares numpy to onnxruntime on simple functions

  :ref:`sphx_glr_gyexamples_plot_benchmark_onnx_function.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Compares numpy to onnxruntime on simple functions</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This short code compares the execution of the operator *Slice* between numpy and onnxruntime fo...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_benchmark_op_thumb.png
    :alt: Benchmark and profile of operator Slice

  :ref:`sphx_glr_gyexamples_plot_benchmark_op.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Benchmark and profile of operator Slice</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="l-custom-parser shows how to add a parser to define a converter a model which works differently...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_mcustom_parser_dataframe_thumb.png
    :alt: Add a parser to handle dataframes

  :ref:`sphx_glr_gyexamples_plot_mcustom_parser_dataframe.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Add a parser to handle dataframes</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Python implements multithreading but it is not working in practice due to the GIL (see Le GIL)....">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_parallel_execution_thumb.png
    :alt: Multithreading with onnxruntime

  :ref:`sphx_glr_gyexamples_plot_parallel_execution.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Multithreading with onnxruntime</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This example follows the same steps introduced in example l-orttraining-linreg-cpu but on GPU. ...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_orttraining_linear_regression_gpu_thumb.png
    :alt: Train a linear regression with onnxruntime-training on GPU in details

  :ref:`sphx_glr_gyexamples_plot_orttraining_linear_regression_gpu.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Train a linear regression with onnxruntime-training on GPU in details</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="epkg:`pytorch` or tensorflow usually work faster if the deep learning model is entirely run out...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_benchmark_eager_mode_thumb.png
    :alt: Benchmark onnxruntime API: eager mode

  :ref:`sphx_glr_gyexamples_plot_benchmark_eager_mode.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Benchmark onnxruntime API: eager mode</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="onnxruntime-training only computes the gradient values. A gradient descent can then use it to t...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_orttraining_linear_regression_cpu_thumb.png
    :alt: Train a linear regression with onnxruntime-training in details

  :ref:`sphx_glr_gyexamples_plot_orttraining_linear_regression_cpu.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Train a linear regression with onnxruntime-training in details</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Example l-plot-parallel-execution shows that parallelizing the inference over multiple GPUs on ...">

.. only:: html

  .. image:: /gyexamples/images/thumb/sphx_glr_plot_parallel_execution_big_model_thumb.png
    :alt: Multithreading with onnxruntime and big models

  :ref:`sphx_glr_gyexamples_plot_parallel_execution_big_model.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Multithreading with onnxruntime and big models</div>
    </div>


.. raw:: html

    </div>


.. toctree::
   :hidden:

   /gyexamples/plot_qextend_onnxruntime
   /gyexamples/plot_benchmark_inference_standard
   /gyexamples/plot_cbegin_opset
   /gyexamples/plot_abegin_convert_pipeline
   /gyexamples/plot_gexternal_lightgbm
   /gyexamples/plot_gbegin_cst
   /gyexamples/plot_fbegin_investigate
   /gyexamples/plot_dbegin_options_list
   /gyexamples/plot_orttraining_nn_gpu_fwbw
   /gyexamples/plot_orttraining_nn_gpu
   /gyexamples/plot_woe_transformer
   /gyexamples/plot_gconverting
   /gyexamples/plot_orttraining_linear_regression_fwbw
   /gyexamples/plot_bbegin_measure_time
   /gyexamples/plot_gbegin_dataframe
   /gyexamples/plot_funny_sigmoid
   /gyexamples/plot_orttraining_linear_regression
   /gyexamples/plot_benchmark_onnx_serialize
   /gyexamples/plot_orttraining_nn_gpu_fwbw_nesterov
   /gyexamples/plot_kcustom_converter_wrapper
   /gyexamples/plot_gexternal_lightgbm_reg_per
   /gyexamples/plot_mcustom_parser
   /gyexamples/plot_profile_ort
   /gyexamples/plot_dbegin_options
   /gyexamples/plot_jcustom_syntax
   /gyexamples/plot_dbegin_options_zipmap
   /gyexamples/plot_lcustom_options
   /gyexamples/plot_gbegin_transfer_learning
   /gyexamples/plot_gexternal_xgboost
   /gyexamples/plot_profile_ort_onnx
   /gyexamples/plot_gexternal_lightgbm_reg_mono
   /gyexamples/plot_gexternal_lightgbm_reg
   /gyexamples/plot_icustom_converter
   /gyexamples/plot_ebegin_float_double
   /gyexamples/plot_benchmark_graph_opt
   /gyexamples/plot_orttraining_benchmark_fwbw
   /gyexamples/plot_orttraining_benchmark
   /gyexamples/plot_benchmark_op_leakyrelu
   /gyexamples/plot_transformer_discrepancy
   /gyexamples/plot_benchmark_op_short
   /gyexamples/plot_catwoe_transformer
   /gyexamples/plot_quantization
   /gyexamples/plot_benchmark_ort_api
   /gyexamples/plot_orttraining_benchmark_torch
   /gyexamples/plot_orttraining_benchmark_fwbw_cls
   /gyexamples/plot_usparse_xgboost
   /gyexamples/plot_benchmark_inference
   /gyexamples/plot_pextend_python_runtime
   /gyexamples/plot_benchmark_onnx_function
   /gyexamples/plot_benchmark_op
   /gyexamples/plot_mcustom_parser_dataframe
   /gyexamples/plot_parallel_execution
   /gyexamples/plot_orttraining_linear_regression_gpu
   /gyexamples/plot_benchmark_eager_mode
   /gyexamples/plot_orttraining_linear_regression_cpu
   /gyexamples/plot_parallel_execution_big_model


.. only:: html

  .. container:: sphx-glr-footer sphx-glr-footer-gallery

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

      :download:`Download all examples in Python source code: gyexamples_python.zip </gyexamples/gyexamples_python.zip>`

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

      :download:`Download all examples in Jupyter notebooks: gyexamples_jupyter.zip </gyexamples/gyexamples_jupyter.zip>`


.. only:: html

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

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