dearborn-open-ai / onedal

oneAPI Data Analytics Library (oneDAL)

oneAPI Data Analytics Library

Installation   |   Documentation   |   Support   |   Examples   |   How to Contribute   

oneAPI Data Analytics Library (oneDAL) is a powerful machine learning library that helps you accelerate big data analysis at all stages: preprocessing, transformation, analysis, modeling, validation, and decision making.

The library implements classical machine learning algorithms. The boost in their performance is achieved by leveraging the capabilities of Intel® hardware.

oneDAL is part of oneAPI. The current branch implements version 1.1 of oneAPI Specification.

Usage

There are different ways for you to build high-performance data science applications that use the advantages of oneDAL:

• Use oneDAL C++ interfaces with or without SYCL support (learn more)
• Use Intel(R) Extension for Scikit-learn* to accelerate existing scikit-learn code without changing it
• Use daal4py, a standalone package with Python API for oneDAL Deprecation Notice: The Java interfaces are removed from the oneDAL library.

Installation

Check System Requirements before installing oneDAL.

You can download the specific version of oneDAL or install it from sources.

Examples

C++ Examples:

• oneAPI interfaces with SYCL support
• oneAPI interfaces without SYCL support
• DAAL interfaces

Python Examples:

• scikit-learn-intelex
• daal4py

Other Examples

• MPI
• MySQL

Documentation

oneDAL documentation:

• Release Notes
• Get Started Guide
• Developer Guide and Reference

Other related documentation:

• daal4py documentation
• Intel(R) Extension for Scikit-learn* documentation
• oneDAL Specifications

Apache Spark MLlib

oneDAL library is used for Spark MLlib acceleration as part of OAP MLlib project and allows you to get a 3-18x increase in performance compared to the default Apache Spark MLlib.

Technical details: FPType: double; HW: 7 x m5.2xlarge AWS instances; SW: Intel DAAL 2020 Gold, Apache Spark 2.4.4, emr-5.27.0; Spark config num executors 12, executor cores 8, executor memory 19GB, task cpus 8

Scaling

oneDAL supports distributed computation mode that shows excellent results for strong and weak scaling:

oneDAL K-Means fit, strong scaling result	oneDAL K-Means fit, weak scaling results

Technical details: FPType: float32; HW: Intel Xeon Processor E5-2698 v3 @2.3GHz, 2 sockets, 16 cores per socket; SW: Intel® DAAL (2019.3), MPI4Py (3.0.0), Intel® Distribution Of Python (IDP) 3.6.8; Details available in the article https://arxiv.org/abs/1909.11822

Support

Ask questions and engage in discussions with oneDAL developers, contributers, and other users through the following channels:

• GitHub Discussions
• Community Forum

You may reach out to project maintainers privately at [email protected].

Security

To report a vulnerability, refer to Intel vulnerability reporting policy.

Contribute

We welcome community contributions. Check our contributing guidelines to learn more.

License

oneDAL is distributed under the Apache License 2.0 license. See LICENSE for more information.

oneMKL FPK microlibs are distributed under Intel Simplified Software License. Refer to third-party-programs-mkl.txt for details.