mirror of
https://github.com/saymrwulf/onnxruntime.git
synced 2026-05-15 20:50:42 +00:00
a6ce7b339f
### Description This PR enables execution of subgraphs in OVEP and currently, when OVEP developers install the onnxruntime-openvino package on windows from pypi, they would have to additionally download OpenVINO windows binaries and run the setupvars.bat script which sets the environment PATH to locate the OV dll's. Also this PR fixes issues of OVEP windows io buffer sample. ### Motivation and Context Fix: We want to make the user experience easy for OVEP Python developers on windows platform. This fix, introduces a function add_openvino_libs_to_path at the location tools/python/util/add_openvino_win_libs.py. The above function, can be called by OVEP python users in the application code and that takes care of setting the OpenVINO dll's to the path from the OpenVINO pypi packge (openvino) which was installed. This change also makes sure that add_openvino_libs_to_path() function is added to onnxruntime python package only when it is build for OpenVINO Execution Provider for ONNXRuntime and not for default ORT python package builds. New user experience for Python OVEP developers on windows platform: step 1: pip install onnxruntime-openvino step 2: pip install openvino step 3: <Add these 2 lines in the application code> import onnxruntime.tools.add_openvino_win_libs as utils utils.add_openvino_libs_to_path() --------- Signed-off-by: MaajidKhan <n.maajid.khan@intel.com> Co-authored-by: MaajidKhan <n.maajid.khan@intel.com> Co-authored-by: Suryaprakash Shanmugam <suryaprakash.shanmugam@intel.com>
67 lines
3.5 KiB
ReStructuredText
67 lines
3.5 KiB
ReStructuredText
OpenVINO™ Execution Provider for ONNX Runtime
|
|
===============================================
|
|
|
|
`OpenVINO™ Execution Provider for ONNX Runtime <https://onnxruntime.ai/docs/execution-providers/OpenVINO-ExecutionProvider.html>`_ is a product designed for ONNX Runtime developers who want to get started with OpenVINO™ in their inferencing applications. This product delivers `OpenVINO™ <https://software.intel.com/content/www/us/en/develop/tools/openvino-toolkit.html>`_ inline optimizations which enhance inferencing performance with minimal code modifications.
|
|
|
|
OpenVINO™ Execution Provider for ONNX Runtime accelerates inference across many `AI models <https://github.com/onnx/models>`_ on a variety of Intel® hardware such as:
|
|
- Intel® CPUs
|
|
- Intel® integrated GPUs
|
|
- Intel® discrete GPUs
|
|
|
|
Installation
|
|
------------
|
|
|
|
Requirements
|
|
^^^^^^^^^^^^
|
|
|
|
- Ubuntu 18.04, 20.04, RHEL(CPU only) or Windows 10 - 64 bit
|
|
- Python 3.8, 3.9 or 3.10 for Linux and only Python3.10 for Windows
|
|
|
|
This package supports:
|
|
- Intel® CPUs
|
|
- Intel® integrated GPUs
|
|
|
|
``pip3 install onnxruntime-openvino``
|
|
|
|
Please install OpenVINO™ PyPi Package separately for Windows.
|
|
For installation instructions on Windows please refer to `OpenVINO™ Execution Provider for ONNX Runtime for Windows <https://github.com/intel/onnxruntime/releases/>`_.
|
|
|
|
**OpenVINO™ Execution Provider for ONNX Runtime** Linux Wheels comes with pre-built libraries of OpenVINO™ version 2023.0.0 eliminating the need to install OpenVINO™ separately. The OpenVINO™ libraries are prebuilt with CXX11_ABI flag set to 0.
|
|
|
|
The package also includes module that is used by torch-ort-inference to accelerate inference for PyTorch models with OpenVINO Execution Provider.
|
|
See `torch-ort-inference <https://github.com/pytorch/ort#accelerate-inference-for-pytorch-models-with-onnx-runtime-preview>`_ for more details.
|
|
|
|
For more details on build and installation please refer to `Build <https://onnxruntime.ai/docs/build/eps.html#openvino>`_.
|
|
|
|
Usage
|
|
^^^^^
|
|
|
|
By default, Intel® CPU is used to run inference. However, you can change the default option to either Intel® integrated or discrete GPU.
|
|
Invoke `the provider config device type argument <https://onnxruntime.ai/docs/execution-providers/OpenVINO-ExecutionProvider.html#summary-of-options>`_ to change the hardware on which inferencing is done.
|
|
|
|
For more API calls and environment variables, see `Usage <https://onnxruntime.ai/docs/execution-providers/OpenVINO-ExecutionProvider.html#configuration-options>`_.
|
|
|
|
Samples
|
|
^^^^^^^^
|
|
|
|
To see what you can do with **OpenVINO™ Execution Provider for ONNX Runtime**, explore the demos located in the `Examples <https://github.com/microsoft/onnxruntime-inference-examples/tree/main/python/OpenVINO_EP>`_.
|
|
|
|
License
|
|
^^^^^^^^
|
|
|
|
**OpenVINO™ Execution Provider for ONNX Runtime** is licensed under `MIT <https://github.com/microsoft/onnxruntime/blob/main/LICENSE>`_.
|
|
By contributing to the project, you agree to the license and copyright terms therein
|
|
and release your contribution under these terms.
|
|
|
|
Support
|
|
^^^^^^^^
|
|
|
|
Please submit your questions, feature requests and bug reports via `GitHub Issues <https://github.com/microsoft/onnxruntime/issues>`_.
|
|
|
|
How to Contribute
|
|
^^^^^^^^^^^^^^^^^^
|
|
|
|
We welcome community contributions to **OpenVINO™ Execution Provider for ONNX Runtime**. If you have an idea for improvement:
|
|
|
|
* Share your proposal via `GitHub Issues <https://github.com/microsoft/onnxruntime/issues>`_.
|
|
* Submit a `Pull Request <https://github.com/microsoft/onnxruntime/pulls>`_.
|