Skip to content
This repository has been archived by the owner on Dec 5, 2024. It is now read-only.

includes minor changes to README and documentation #101

Merged
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,7 @@
[![PyPI version](https://badge.fury.io/py/pgmax.svg)](https://badge.fury.io/py/pgmax)
[![pre-commit.ci status](https://results.pre-commit.ci/badge/github/vicariousinc/PGMax/master.svg)](https://results.pre-commit.ci/latest/github/vicariousinc/PGMax/master)
[![codecov](https://codecov.io/gh/vicariousinc/PGMax/branch/master/graph/badge.svg?token=FrRlTDCFjk)](https://codecov.io/gh/vicariousinc/PGMax)
[![Documentation Status](https://readthedocs.org/projects/pgmax/badge/?version=latest)](https://pgmax.readthedocs.io/en/latest/?badge=latest)

# PGMax

Expand Down
7 changes: 6 additions & 1 deletion docs/source/index.rst
Original file line number Diff line number Diff line change
@@ -1,6 +1,11 @@
PGMax Reference Documentation
==============================
PGMax is a library for working with Factor Graphs in `JAX <https://jax.readthedocs.io/en/latest/>`_. It currently provides an interface for specifying factor graphs of any type, as well as an efficient implementation of max-product belief propagation and inference on these graphs.
PGMax implements general factor graphs for probabilistic graphical models (PGMs) with discrete variables, and hardware-accelerated differentiable loopy belief propagation (LBP) in `JAX <https://jax.readthedocs.io/en/latest/>`_.

- General factor graphs: PGMax goes beyond pairwise PGMs, and supports arbitrary factor graph topology, including higher-order factors.
- LBP in JAX: PGMax generates pure JAX functions implementing LBP for a given factor graph. The generated pure JAX functions run on modern accelerators (GPU/TPU), work with JAX transformations (e.g. ``vmap`` for processing batches of models/samples, ``grad`` for differentiating through the LBP iterative process), and can be easily used as part of a larger end-to-end differentiable system.



.. toctree::
:maxdepth: 1
Expand Down
2 changes: 2 additions & 0 deletions pyproject.toml
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,7 @@ name = "pgmax"
version = "0.2.0"
description = "Loopy belief propagation for factor graphs on discrete variables, in JAX!"
authors = ["Stannis Zhou <stannis@vicarious.com>", "Nishanth Kumar <njk@csail.mit.edu>", "Miguel Lazaro-Gredilla <miguel@vicarious.com>", "Dileep George <dileep@vicarious.com>"]
maintainers = ["Stannis Zhou <stannis@vicarious.com>", "Nishanth Kumar <njk@csail.mit.edu>"]
# Additional info
license = "MIT"
readme = "README.md"
Expand All @@ -11,6 +12,7 @@ include = [
]
homepage = "https://github.com/vicariousinc/PGMax"
repository = "https://github.com/vicariousinc/PGMax"
documentation = "https://pgmax.readthedocs.io/en/latest/"
keywords = ["PGMax", "jax"]

[tool.poetry.dependencies]
Expand Down