SheffieldML/GPy

Gaussian processes framework in python

What it solves

GPy is a framework for implementing and using Gaussian processes, providing a structured way to handle regression and other GP-based modeling tasks in Python.

How it works

It provides a set of models and kernels to perform Gaussian process regression. The framework separates the core parameterization and gradient-based model optimization into a separate package called paramz.

Who it’s for

Researchers and developers who need a flexible Python-based framework for Gaussian processes, particularly for regression and multi-output/coregionalized regression.

Highlights

  • Gaussian Process Framework: Dedicated tools for GP regression and coregionalized regression.
  • Modular Optimization: Uses the paramz package for pure gradient-based model optimization.
  • Cross-Version Model Saving: Provides a specific method for saving model parameters via NumPy arrays to avoid pickle inconsistency across Python versions.
  • Broad Platform Support: Compatible with Windows, macOS, and Linux on Python 3.9+.

Related

  • Project
    QuantClimate/GPJax

    A low-level Gaussian process framework built in Jax that provides researchers with a flexible, mathematically-aligned interface for building and extending GP models.

  • Project
    GPflow/GPflow

    A Python library for building Gaussian process models using TensorFlow and TensorFlow Probability for fast, GPU-accelerated inference.

  • Project
    cornellius-gp/gpytorch

    A Gaussian process library implemented in PyTorch that enables the creation of scalable, flexible, and GPU-accelerated GP models using numerical linear algebra.

  • Project
    dfm/tinygp

    A lightweight Gaussian Process library built on JAX that supports GPU acceleration and automatic differentiation.

  • Project
    pgmpy/pgmpy

    A Python library for causal and probabilistic reasoning using graphical models, providing tools for structure learning, parameter estimation, and causal inference.