onnx/sklearn-onnx

Convert scikit-learn models and pipelines to ONNX

What it solves

It provides a way to convert scikit-learn models into the Open Neural Network Exchange (ONNX) format. This allows users to move models from the training environment to a high-performance scoring environment using tools like ONNX Runtime, improving inference speed and portability.

How it works

The library converts scikit-learn models and transformers into ONNX operators. It supports a wide range of supported scikit-learn models and allows for the registration of external converters to handle models or transformers from other libraries within a scikit-learn pipeline.

Who it’s for

Data scientists and ML engineers who train models using scikit-learn but need to deploy them in production environments that require high-performance inference engines.

Highlights

  • Converts scikit-learn models to ONNX format.
  • Compatible with ONNX Runtime for high-performance scoring.
  • Supports external converter registration for custom pipelines.
  • Supports the latest ONNX opset 21.

Related

  • Project
    onnx/models

    A curated collection of pre-trained, state-of-the-art machine learning models in the ONNX format, enabling cross-framework compatibility and deployment.

  • Project
    microsoft/onnxconverter-common

    A collection of common utilities and functions used by converters to translate AI models from various frameworks into the ONNX format.

  • Project
    onnx/onnxmltools

    A toolkit for converting machine learning models from various frameworks like TensorFlow, scikit-learn, and XGBoost into the ONNX format for cross-platform deployment.

  • Project
    Talmaj/onnx2pytorch

    A library that transforms ONNX models into PyTorch models, allowing users to utilize PyTorch's backend and neural network manipulation tools.

  • Project
    microsoft/onnxruntime

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