Intel® Extension for TensorFlow* is a heterogeneous, high performance deep learning extension plugin based on TensorFlow PluggableDevice interface, aiming to bring Intel CPU or GPU devices into TensorFlow open source community for AI workload acceleration. It allows users to flexibly plug an XPU into TensorFlow on-demand, exposing the computing power inside Intel's hardware.
This diagram provides a summary of the TensorFlow* PyPI package ecosystem.
-
TensorFlow PyPI packages: estimator, keras, tensorboard, tensorflow-base
-
Intel® Extension for TensorFlow* package:
intel_extension_for_tensorflow
contains:- XPU specific implementation
- Kernels & operators
- Graph optimizer
- Device runtime
- XPU configuration management
- XPU backend selection
- Options turning on/off advanced features
- XPU specific implementation
Intel® Extension for TensorFlow* provides Intel XPU and Intel CPU support.
Package | CPU | XPU | Installation |
---|---|---|---|
Intel GPU driver | Y | Install Intel GPU driver | |
Intel® oneAPI Base Toolkit | Y | Install Intel® oneAPI Base Toolkit | |
TensorFlow | Y | Y | Install TensorFlow 2.15.0 |
Intel® Extension for TensorFlow* can be installed through the following channels:
- PyPI: XPU \ CPU
- DockerHub: XPU Container \ CPU Container
- Source: Build from source
Intel® Extension for TensorFlow* | Stock TensorFlow |
---|---|
Not support | 2.17 |
latest build from source | 2.16 |
v2.15.0.0 & v2.15.0.1 | 2.15 |
v2.14.0.1 & v2.14.0.2 | 2.14 |
v2.13.0.0 | 2.13 |
v1.2.0 | 2.12 |
v1.1.0 | 2.10 & 2.11 |
v1.0.0 | 2.10 |
pip install --upgrade intel-extension-for-tensorflow[xpu]
Environment check instructions for XPU:
export path_to_site_packages=`python -c "import site; print(site.getsitepackages()[0])"`
bash ${path_to_site_packages}/intel_extension_for_tensorflow/tools/env_check.sh
Refer to XPU installation for details.
pip install --upgrade intel-extension-for-tensorflow[cpu]
Sanity check instructions:
python -c "import intel_extension_for_tensorflow as itex; print(itex.__version__)"
pip install --upgrade intel-extension-for-tensorflow-weekly[xpu] -f https://developer.intel.com/itex-whl-weekly
Environment check instructions for GPU weekly:
export path_to_site_packages=`python -c "import site; print(site.getsitepackages()[0])"`
bash ${path_to_site_packages}/intel_extension_for_tensorflow/tools/env_check.sh
pip install --upgrade intel-extension-for-tensorflow-weekly[cpu] -f https://developer.intel.com/itex-whl-weekly
Sanity check instructions:
python -c "import intel_extension_for_tensorflow as itex; print(itex.__version__)"
Visit the online document website, and then get started with a tour of Intel® Extension for TensorFlow* examples.
We welcome community contributions to Intel® Extension for TensorFlow*.
This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the Contributor Covenant. Please see contribution guidelines for additional details.
- TensorFlow GPU device plugins
- Accelerating TensorFlow on Intel® Data Center GPU Flex Series
- Meet the Innovation of Intel AI Software: Intel® Extension for TensorFlow*
- Efficient TensorFlow Distributed Training on Intel Data Center GPU Max Series
- Accelerate JAX models on Intel GPUs via PJRT
- Running TensorFlow Stable Diffusion on Intel Arc GPUs
- AI workload Acceleration with Intel® Extension for TensorFlow* | Intel Software
Submit your questions, feature requests, and bug reports on the GitHub issues page.
See Intel's Security Center for information on how to report a potential security issue or vulnerability.
See also: Security Policy
This distribution includes third party software governed by separate license terms. This third party software, even if included with the distribution of the Intel software, may be governed by separate license terms, including without limitation, third party license terms, other Intel software license terms, and open source software license terms. These separate license terms govern your use of the third party programs as set forth in the "THIRD-PARTY-PROGRAMS" file.