Intel® Distribution for Python*

Develop fast, performant Python code with this set of essential computational packages including NumPy, SciPy, scikit-learn*, and more.

Achieve Greater Performance with Minimal or No Code Changes

  • Near-native performance comes through acceleration of core Python numerical packages. This is done using 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 apply 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.
  • Program for multiple devices using the same programming model, DPPy (Data Parallel Python), thereby avoiding rewriting CPU code to device code. Its included package, dpCtl, lets you control execution and data management across the host and devices.
  • Run your numeric and machine learning (ML) workloads on data-parallel and heterogeneous hardware without having to go out of the Python ecosystem, while simultaneously not compromising on performance using DPPy.

Develop in the Cloud

Get what you need to build and optimize your oneAPI projects for free. With an Intel® DevCloud account, you get 120 days of access to the latest Intel® hardware—CPUs, GPUs, FPGAs—and Intel® oneAPI tools and frameworks. No software downloads. No configuration steps. No installations.

Get Access

 

Download the Library

Intel Distribution for Python is included as part of the Intel® AI Analytics Toolkit.

Get It Now

 

Who Needs This Product

Machine Learning Developers, Data Scientists, and Analysts

Implement performance-packed, production-ready scikit-learn 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 speed up your Python applications.

What's New in 2021?

This release offers many new features and improvements, including:

  • Data Parallel Python (DPPy), a set of packages essential for Python development. DPPy includes dpCtl, the package for controlling execution on multiple devices and for data management, and Data Parallel Numeric Python (dpnp), a device-accelerated package compatible with dpCtl.
    dpCtl documentation | dpnp documentation
  • Faster machine learning with XGBoost*, scikit-learn, and advanced ML usages, including multiple devices, with daal4py.
  • Scikit-ipp for image warping, image filtering, and morphological operations. Includes support for transform function multithreading and partial multithreading for filters using OpenMP.
  • An updated version of NumPy.
  • New dpnp package for array computations on SYCL devices.
  • Improved Numba compiler to accelerate custom Python code targeted to CPU and GPU execution.

For a complete and up-to-date list, see the release notes.

Get the Single Component

A stand-alone version of this component is available.

Download

Download from Package Managers or Containers

Conda | YUM | APT | Docker

Give Feedback