https://github.com/probml/pgm-jax
Probabilistic Graphical Models in JAX
https://github.com/probml/pgm-jax
Last synced: about 1 year ago
JSON representation
Probabilistic Graphical Models in JAX
- Host: GitHub
- URL: https://github.com/probml/pgm-jax
- Owner: probml
- License: mit
- Created: 2022-04-12T00:06:58.000Z (about 4 years ago)
- Default Branch: main
- Last Pushed: 2022-10-11T17:45:44.000Z (over 3 years ago)
- Last Synced: 2025-03-28T10:47:36.560Z (over 1 year ago)
- Language: Jupyter Notebook
- Homepage:
- Size: 186 KB
- Stars: 14
- Watchers: 6
- Forks: 2
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# pgm-jax
Probabilistic Graphical Models in JAX.
This is a work in progress.
Currently we only support belief propagation in (loopy)
Gaussian PGMs. In the case of chain structured graphs
with linear-Gaussian potentials,
this gives the same result as Kalman smoothing.