Quick Get Started¶

Intel® Extension for TensorFlow* is a heterogeneous, high performance deep learning extension plugin based on TensorFlow PluggableDevice interface to bring Intel CPU or GPU devices into TensorFlow open source community for AI workload acceleration. It allows flexibly plugging an XPU into TensorFlow on-demand, and exposing 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

Install¶

Hardware Requirement¶

Intel® Extension for TensorFlow* provides Intel GPU support and experimental Intel CPU support.

Software Requirement¶

Package	CPU	GPU	Installation
Intel GPU driver		Y	Install Intel GPU driver
Intel® oneAPI Base Toolkit		Y	Install Intel® oneAPI Base Toolkit
TensorFlow	Y	Y	Install TensorFlow 2.12.0

Installation Channel:¶

Intel® Extension for TensorFlow* can be installed from the following channels:

PyPI: GPU \ CPU
DockerHub: GPU Container \ CPU Container
Source: Build from source

Compatibility Table¶

Intel® Extension for TensorFlow*	Stock TensorFlow
v1.2.0	2.12
v1.1.0	2.10 & 2.11
v1.0.0	2.10

Install for GPU¶

pip install tensorflow==2.12
pip install intel-extension-for-tensorflow[gpu]==1.2.0

Refer to GPU installation for details.

Install for CPU [Experimental]¶

pip install tensorflow==2.12
pip install intel-extension-for-tensorflow[cpu]==1.2.0

Sanity check by:

python -c "import intel_extension_for_tensorflow as itex; print(itex.__version__)"

Resources¶

Support¶

Submit your questions, feature requests, and bug reports on the GitHub issues page.