google-deepmind/jraph
A library for building Graph Neural Networks using JAX, developed by Google DeepMind.
Not currently ranked — collecting fresh signals.
star history
Jraph provides data structures for representing graphs, utilities for graph manipulation, and a collection of pre-built GNN models. It leverages JAX’s functional transformations for flexible neural network construction and supports distributed training for large-scale graph networks with explicit edge messaging across multiple devices.
Frequently asked
- What is google-deepmind/jraph?
- A library for building Graph Neural Networks using JAX, developed by Google DeepMind.
- Is jraph open source?
- Yes — google-deepmind/jraph is open source, released under the Apache-2.0 license.
- What language is jraph written in?
- google-deepmind/jraph is primarily written in Python.
- How popular is jraph?
- google-deepmind/jraph has 1.5k stars on GitHub.
- Where can I find jraph?
- google-deepmind/jraph is on GitHub at https://github.com/google-deepmind/jraph.