Education Ecosystem Blog

It should be no surprise that Python continues to grow in popularity. Data Scientists, Machine Learning (ML) developers and all manner of data junkies love the ease of creating Python code – but many are put off by the slow execution that’s inherent with most interpreted languages like Python.

It was with one goal – accelerating Python execution performance – that lead to the creation of Intel® Distribution for Python, a set of tools that let anyone speed Python application performance right out of the box, usually with no code changes required.

Intel® Distribution for Python* speeds NumPy, SciPy, and scikit-learn by integrating the Intel® Math Kernel Library (Intel® MKL) and Intel® Data Analytics Acceleration Library (Intel® DAAL) both written in C and assembler to speed up Python math functions.

Also, Intel® Distribution for Python* incorporates the latest advances in vectorization and threading, Numba and Cython to deliver composable parallelism with Threaded Building Blocks (TBB). The boost to high-performance computing (HPC) applications and other Python code can be dramatic. Intel has documented performance increases up to 200 times faster and more – that’s 200x, not 200%.

Anyone using Python can benefit from this product, from ML developers, to data scientists, numerical and scientific computing developers, and HPC developers to name just a few.

The latest release features faster ML with scikit-learn key algorithms accelerated with Intel DAAL, the latest TensorFlow and Caffe libraries optimized for Intel architecture, and the XGBoost package included (in the Linux version).

All your Python runs, just a lot faster that it did on regular opensource Python.

And, if you need help, there is a community forum where you can connect to the develop community and Intel technical experts eager to lend a hand. Priority support is also available if needed.

Intel® Distribution for Python* will run on Linux, Windows 10, and macOS systems, with support for Python versions 2.7 and 3.6, and is compatible with Microsoft Visual Studio and PyCharm.

Intel is offering this product for free download here. You can also download the entire package from your favorite open-source community, including Conda pacakages, Pip Installer, a Docker image, Amazin Machine Images (AMI), and the YUM and APT Linux repositories.

Why wait? Your Python code isn’t going to accelerate itself.