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

Corner plots, LaTeX tables and plotting walks.

Combination of transformers and diffusion models for flexible all-in-one simulation-based inference

Simulation platform for general-purpose robotics & embodied AI learning.

A repository to store state of the art score model architectures

High accuracy RAG for answering questions from scientific documents with citations

Simple OOP Python Code to run some Pseudo-Spectral 2D Simulations of Turbulence

Generates vertical profiles of birds from weather radar volume scans

A modular graph-based Retrieval-Augmented Generation (RAG) system

Multimodal contrastive pretraining for astronomical data

E(n) Equivariant GNN in jax

Python bindings and scikit-learn interface for the Operon library for symbolic regression.

Physical Symbolic Optimization

Turn SymPy expressions into trainable JAX expressions.

Information Maximizing Neural Network compression of the cosmic 21-cm signal

Official implementation for the paper: Permutation Invariant Graph Generation via Score-Based Generative Modeling

Graph neural networks in JAX.

Test bench for galaxy and halo bias models

3D U-Net model for volumetric semantic segmentation written in pytorch

A lightweight PyTorch implementation of the Transformer-XL architecture proposed by Dai et al. (2019)

Coverage tests to check the quality of your posterior estimators.

Curated list of interactive ML demos

Scalable graph analytics database powered by a multithreaded, vectorized temporal engine, written in Rust

MT3: Multi-Task Multitrack Music Transcription

Convert Machine Learning Code Between Frameworks

Transformer-XL for Jazz music composition. Paper: "The Jazz Transformer on the Front Line: Exploring the Shortcomings of AI-Composed Music through Quantitative Measures", ISMIR 2020