Metadata-Version: 2.1
Name: veroku
Version: 0.1.82
Summary: An open source library for building and performing inference with probabilistic graphical models.
Home-page: https://github.com/ejlouw/veroku
Author: EJ Louw
Author-email: ejlouw00@gmail.com
License: BSD 3-Clause License
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: BSD License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
Requires-Dist: Sphinx (==3.2.1)
Requires-Dist: autodoc (==0.5.0)
Requires-Dist: matplotlib (==3.0.3)
Requires-Dist: seaborn (==0.11.0)
Requires-Dist: graphviz (==0.10.1)
Requires-Dist: Pillow (==7.1.0)
Requires-Dist: importnb (==0.6.2)
Requires-Dist: ipython (==7.5.0)
Requires-Dist: pandas (==1.1.3)
Requires-Dist: tqdm (==4.48.1)
Requires-Dist: ipympl (==0.5.8)
Requires-Dist: numpy (==1.19.2)
Requires-Dist: networkx (==2.5)
Requires-Dist: numdifftools (==0.9.20)
Requires-Dist: scipy (==1.5.2)
Requires-Dist: mockito (==1.2.2)

<div align="center">
  <img src="logo.svg">
</div>

### Disclaimer
This project is still in the very early stages of development. Use this library at your own risk. The github repo is also still private,
but will be made public soon.

### Overview
<div style="text-align: justify">
Veroku is an open source library for building and performing inference with probabilistic graphical models (PGMs). PGMs
provide a framework for performing efficient probabilistic inference with very high dimensional distributions. A typical
example of a well-known type of PGM is the Kalman filter that can be used to obtain probabilistic estimates of a hidden
state of a process or system, given noisy measurements. PGMs can in principle be used for any problem that involves
uncertainty and is therefore applicable to many problems.</div> 

Veroku currently supports the following distributions:
* Multivariate Gaussian
* Multivariate Gaussian mixture
* Multivariate categorical

<div style="text-align: justify">
These distributions can be used as factors to represent a factorised distribution. These factors can be used, together
with the `cluster_graph` module to automatically create valid cluster graphs. Inference can be performed in these graphs
using message passing algorithms. Currently only the LBU (Loopy Belief Update) message-passing algorithm is supported.
</div>

<br/><br/>

### Future Features
To be added soon:
* Example notebooks
* Non-linear Gaussian distribution
* Plate models (for efficiently specifying PGMs as modular/hierarchical templates)

On the roadmap:
* Dirichlet distribution
* Wishart distribution
* Normal-Wishart distribution



