intel/python

Intel® Distribution for Python*

https://www.intel.com/content/www/us/en/developer/tools/oneapi/distribution-for-python.html#gs.9bos9m enhances performance and can improve your program speed from 10 to 100 times faster. It is a Python* distribution that includes the https://www.intel.com/content/www/us/en/developer/tools/oneapi/onemkl.html (oneMKL) and other Intel performance libraries to enable near-native performance through acceleration of core numerical and machine learning packages.

Intel® Distribution for Python* is also available for Intel® dGPUs, that include the latest https://dgpu-docs.intel.com/releases/releases.html and Intel® OneAPI runtime libraries such as https://www.intel.com/content/www/us/en/developer/tools/oneapi/onemkl.html, https://www.intel.com/content/www/us/en/developer/tools/oneapi/dpc-compiler-download.html and https://www.intel.com/content/www/us/en/developer/tools/oneapi/oneccl.html that enable Machine Learning frameworks leverage the XPU device plugin.

Images

The images below include variations for only the core packages in the https://www.intel.com/content/www/us/en/developer/tools/oneapi/distribution-for-python.html#gs.9bos9m installation, or all of the packages.

Run a Performance Sample

To run a performance sample run the following commands:

bash
git clone https://github.com/intel/ai-containers
cd ai-containers/python
docker run --rm -it \
    -v $PWD/tests:/tests \
    intel/python:latest \
    python /tests/perf_sample.py

Compare the results against stock python

In the previous command, you should see a result at the bottom like: Time Consuming: 0.03897857666015625. We can compare this against python:3.11-slim-bullseye

bash
# Use the working directory from the above command
docker run --rm -it \
    -v $PWD/tests:/tests \
    python:3.10-slim-bullseye \
    bash
pip install numpy
python /tests/perf_sample.py

Run a Sanity test using the XPU variant

Use the following command to check the availability of Intel dGPU devices on your system and the presence of Intel® OneAPI runtime libraries.

bash
# Use the working directory from the first command
docker run --rm -it \
    -v $PWD/tests:/tests \
    --device /dev/dri \
    intel/python:3.11-xpu-full \
    bash /tests/xpu_base_layers_test.sh

Build from Source (Advanced)

To build the images from source, clone the https://github.com/intel/ai-containers repository, follow the main README.md file to setup your environment, and run the following command:

bash
cd python
docker compose build idp
docker compose run idp

To build the xpu variant of the image, run the following commands:

bash
cd python
docker compose build xpu
docker compose run xpu

You can find the list of services below for each container in the group:

License

View the https://github.com/intel/ai-containers/blob/main/LICENSE for the [Intel® Distribution for Python].

It is the image user's responsibility to ensure that any use of The images below comply with any relevant licenses for all software contained within.

* Other names and brands may be claimed as the property of others.