Using Intel® Distribution for Python*

You can:

  • Achieve greater application performance with minimal or no code changes. Near-native performance comes through acceleration of core Python numerical packages via Intel® Performance Libraries such as Intel® oneAPI Math Kernel Library (oneMKL) and Intel® oneAPI Data Analytics Library (oneDAL).
  • Compile your custom code using core compiler technologies such as Numba* and Cython. These add vectorization and multithreading to your code bringing Python to new levels.
  • Leverage composable parallelism through Intel® oneAPI Threading Building Blocks (oneTBB) and symmetric multiprocessing (SMP). These technologies help you avoid excess headroom, use nested parallelism without oversubscription, and optimize particular workloads for the highest performance.
  • Run your numeric and ML workloads on data-parallel and heterogeneous hardware without having to go out of the Python ecosystem, while at the same time not compromising on the performance using data parallel Python.

Who Needs This Product

Machine Learning Developers, Data Scientists, and Analysts

Easily implement performance-packed, production-ready, machine learning algorithms.

 

Numerical and Scientific Computing Developers

Accelerate and scale the compute-intensive Python packages NumPy, SciPy, and mpi4py.

 

High-Performance Computing (HPC) Developers

Unlock the power of modern hardware to accelerate your Python applications.

What’s New

This release offers many performance improvements.

  • Faster machine learning with XGBoost* and scikit-learn*. Advanced machine learning usages, including multiple devices, with daal4py
  • Data parallel Python is a set of packages essential for data parallel Python development. It includes dpctl, the package for controlling execution on multiple devices and for data management. Data parallel Python also includes dpnp (data parallel numeric Python), a device-accelerated package compatible with dpctrl
  • Scikit-ipp for image warping, image filtering, and morphological operations. It includes support for multithreading of transform functions and partial multithreading for filters using OpenMP*
  • An improved Numba* compiler to accelerate custom Python code targeted to CPU and GPU execution
  • Scalable Dataframe Compiler (SDC)—the extension of Numba that accelerates pandas operations

Release Notes

Available as Part of the Intel® AI Analytics Toolkit

This toolkit includes high-performing, Intel-optimized Python libraries designed to streamline end-to-end data science and machine learning workflows. Get Intel’s latest analytics and AI optimizations in one place to ensure your software works seamlessly together and with other Intel® oneAPI Toolkits.

Learn More

Download

 

Support

Community Forum

Connect to the developer community and our technical experts through this public forum.

For additional help, see our general oneAPI Support.

Specifications

Target operating systems: Linux*, Windows® 10 , macOS*

Download: Windows, Linux, macOS

Related Tools

Intel® AI Analytics Toolkit

Accelerate end-to-end data analytics and machine learning workloads using this toolkit.

Dejar comentario

Información sobre productos y desempeño

1

El desempeño varía según el uso, la configuración y otros factores. Más información en www.Intel.com/PerformanceIndex.