Skip to content

Latest commit

 

History

History
105 lines (63 loc) · 6.73 KB

INSTALL.md

File metadata and controls

105 lines (63 loc) · 6.73 KB

Installation using Pre-Built Packages

OpenVINO™ integration with TensorFlow is released for Linux, macOS, and Windows. You can choose one of the following methods based on your requirements.

Linux

OpenVINO™ integration with TensorFlow on Linux is released in two different versions: one built with CXX11_ABI=0 and the other built with CXX11_ABI=1.

Since TensorFlow packages available in PyPi are built with CXX11_ABI=0 and OpenVINO™ release packages are built with CXX11_ABI=1, binary releases of these packages cannot be installed together.

  • Includes pre-built libraries of OpenVINO™ version 2022.1.0. The users do not have to install OpenVINO™ separately
  • Supports Intel® CPUs, Intel® integrated GPUs, and Intel® Movidius™ Vision Processing Units (VPUs). No VAD-M support
  • Build with CXX11_ABI=0

  • Compatible with OpenVINO™ version 2022.1.0
  • Supports Intel® CPUs, Intel® integrated GPUs, Intel® Movidius™ Vision Processing Units (VPUs), and Intel® Vision Accelerator Design with Movidius™ (VAD-M)
  • Build with CXX11_ABI=1
  • Needs a custom TensorFlow ABI1 package, which is available in Github release

macOS

  • Includes pre-built libraries of OpenVINO™ version 2022.1.0. The users do not have to install OpenVINO™ separately
  • Supports Intel® CPUs, Intel® integrated GPUs, and Intel® Movidius™ Vision Processing Units (VPUs). No VAD-M support

Windows

  • Includes pre-built libraries of OpenVINO™ version 2022.1.0. The users do not have to install OpenVINO™ separately
  • Supports Intel® CPUs, Intel® integrated GPUs, and Intel® Movidius™ Vision Processing Units (VPUs). No VAD-M support
  • TensorFlow wheel for Windows from PyPi does't have all the API symbols enabled which are required for OpenVINO™ integration with TensorFlow. User needs to install the TensorFlow wheel from the assets of the Github release page.

Pre-built packages summary

TensorFlow Pip Package OpenVINO™ integration with TensorFlow Pip Package Supported OpenVINO™ Flavor Supported Hardwares Comments
tensorflow openvino-tensorflow OpenVINO™ built from source CPU,iGPU,MYRIAD OpenVINO™ libraries are built from source and included in the wheel package
tensorflow-abi1 openvino-tensorflow-abi1 Dynamically links to OpenVINO™ binary release CPU,iGPU,MYRIAD,VAD-M OpenVINO™ integration with TensorFlow libraries are dynamically linked to OpenVINO™ binaries

1.1. Install OpenVINO™ integration with TensorFlow alongside PyPi TensorFlow (Works on Linux, macOS)

    pip3 install -U pip
    pip3 install tensorflow==2.8.0
    pip3 install openvino-tensorflow==2.0.0

The openvino-tensorflow PyPi package is cross-compatible with PATCH versions of TensorFlow. For example, openvino-tensorflow wheel for TF 2.8.0 would work with any future PATCH versions like TF 2.8.1, and 2.8.2.

1.2. Install OpenVINO™ integration with TensorFlow alongside TensorFlow released on Github (Works on Windows)

    pip3.9 install -U pip
    pip3.9 install https://github.com/openvinotoolkit/openvino_tensorflow/releases/download/v2.0.0/tensorflow-2.8.0-cp39-cp39-win_amd64.whl
    pip3.9 install openvino-tensorflow==2.0.0

1.3. Install OpenVINO™ integration with TensorFlow alongside the Intel® Distribution of OpenVINO™ Toolkit (Works on Linux)

  1. Ensure the following versions are being used for pip and numpy:

     pip3 install -U pip
 pip3 install numpy==1.20.2

  2. Install TensorFlow based on your Python version. You can build TensorFlow from source with -D_GLIBCXX_USE_CXX11_ABI=1 or follow the insructions below to use the appropriate package:

     pip3.7 install https://github.com/openvinotoolkit/openvino_tensorflow/releases/download/v2.0.0/tensorflow_abi1-2.8.0-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl

 or

 pip3.8 install https://github.com/openvinotoolkit/openvino_tensorflow/releases/download/v2.0.0/tensorflow_abi1-2.8.0-cp38-cp38-manylinux_2_12_x86_64.manylinux2010_x86_64.whl

 or

 pip3.9 install https://github.com/openvinotoolkit/openvino_tensorflow/releases/download/v2.0.0/tensorflow_abi1-2.8.0-cp39-cp39-manylinux_2_12_x86_64.manylinux2010_x86_64.whl

  3. Download & install Intel® Distribution of OpenVINO™ Toolkit 2022.1.0 release along with its dependencies from (https://software.intel.com/en-us/openvino-toolkit/download).

  4. Initialize the OpenVINO™ environment by running the setupvars.sh located in <openvino_install_directory>/bin using the command below:

     source setupvars.sh

  5. Install openvino-tensorflow. Based on your Python version, choose the appropriate package below:

     pip3.7 install https://github.com/openvinotoolkit/openvino_tensorflow/releases/download/v2.0.0/openvino_tensorflow_abi1-2.0.0-cp37-cp37m-manylinux_2_27_x86_64.whl

 or

 pip3.8 install https://github.com/openvinotoolkit/openvino_tensorflow/releases/download/v2.0.0/openvino_tensorflow_abi1-2.0.0-cp38-cp38-manylinux_2_27_x86_64.whl

 or

 pip3.9 install https://github.com/openvinotoolkit/openvino_tensorflow/releases/download/v2.0.0/openvino_tensorflow_abi1-2.0.0-cp39-cp39-manylinux_2_27_x86_64.whl