jax-ml/oryx

Oryx is a library for probabilistic programming and deep learning built on top of Jax.

What it solves

Oryx provides a framework for probabilistic programming and deep learning, allowing users to build complex statistical models that can bet integrated with JAX's high-performance computing capabilities.

How it works

It implements a set of function transformations that are designed to compose and integrate seamlessly with JAX's core transformations, such as jit (just-in-time compilation), grad (automatic differentiation), and vmap (vectorization).

Who it’s for

Researchers and developers who need to combine probabilistic modeling with the speed and flexibility of JAX.

Highlights

  • Built on top of JAX
  • Supports probabilistic programming
  • Integrates with JAX transformations like jit, grad, and vmap

Related

  • Project
    jax-ml/jax

    A Python library for accelerator-oriented array computation and program transformation, designed for high-performance numerical computing and large-scale machine learning.

  • Project
    tensorflow/probability

    A library for probabilistic reasoning and statistical analysis that integrates probabilistic methods with deep networks using TensorFlow or JAX.

  • Project
    pyro-ppl/numpyro

    A lightweight probabilistic programming library powered by JAX that provides fast Bayesian inference via MCMC and variational inference on GPU/TPU/CPU.

  • Project
    probml/dynamax

    A JAX-based library for probabilistic state space models that provides tools for inference and learning across various model types like HMMs and Gaussian SSMs.

  • Project
    patrick-kidger/equinox

    A JAX library that provides PyTorch-like syntax for building neural networks and models by treating them as PyTrees, ensuring full compatibility with the JAX ecosystem.