uxlfoundation/scikit-learn-intelex

Extension for Scikit-learn is a seamless way to speed up your Scikit-learn application

What it solves

It accelerates machine learning workflows for tabular data by speeding up training and inference in scikit-learn applications. It addresses the performance bottlenecks of standard scikit-learn on various hardware configurations, offering significant speedups without requiring users to change their existing code.

How it works

The extension patches scikit-learn by replacing standard calls with calls to the oneAPI Data Analytics Library (oneDAL). It achieves acceleration through vector instructions, threading, and hardware-specific memory optimizations for both CPUs and GPUs (including multi-GPU and multi-node configurations).

Who it’s for

Data scientists and ML engineers who use scikit-learn for tabular data and want to improve performance on Intel hardware without rewriting their application logic.

Highlights

  • Provides up to 100X acceleration for certain workflows with an average speedup of 8.5x.
  • Supports seamless integration via a patching mechanism that maintains the open-source scikit-learn API.
  • Enables offloading to GPUs and multi-GPU setups for higher performance.
  • Can be used either as a patch to existing scikit-learn imports or as a standalone module.

Related

  • Project
    uxlfoundation/oneDAL

    A C++ and DPC++ library that provides hardware-accelerated machine learning routines for tabular data across CPUs, GPUs, and distributed systems.

  • Project
    NVIDIA/cuml

    A GPU-accelerated machine learning library that provides scikit-learn-style APIs and a transparent acceleration module to speed up ML workloads on NVIDIA GPUs.

  • Project
    h2oai/h2o4gpu

    A collection of GPU-accelerated machine learning solvers that act as a drop-in replacement for scikit-learn, providing Python and R APIs for faster model training.

  • Dispatch
    Intel and Hugging Face Partnership for Machine Learning Hardware Acceleration

    Intel has joined the Hugging Face Hardware Partner Program to develop the Optimum Intel library, enabling streamlined hardware acceleration for training, fine-tuning, and inference of Transformer models on Intel platforms.

  • Project
    microsoft/onnxruntime

    A cross-platform machine learning accelerator that optimizes inference and training for models from various frameworks like PyTorch and TensorFlow.