Metadata-Version: 2.1
Name: pfg
Version: 1.0.0
Summary: A lightweight library for building Factor Graphs and performing inference using the loopy belief propagation algorithm.
Home-page: https://github.com/steveschwarcz/PFG-Python-Factor-Graph-Library
Author: Steven Schwarcz
Author-email: sasz11@gmail.com
License: UNKNOWN
Description: # PFG - The Python Factor Graph Library
        
        `PFG` is a lightweight Python library for building and performing inference
         on Factor Graphs. It is implemented in Python and all operations are
          vectorized, so it runs extremely quickly. 
          
        `PFG` can perform inference using the Belief Propagation algorithm. In the
         case where factor graphs do not have a tree structure, `PFG` can perform the
          Loopy Belief Propagation algorithm, which isn't guaranteed to converge but
           usually gets good results in practice. When performing Loopy belief
            Propagation, `PFG` also allows the user to create a schedule of factors, so
             that in large graphs the user will have complete control over the order
              in which messages are passed during inference.
              
        ## Installation
        
        To install `PFG`, simply run the following command:
        
        ```
        pip install pfg
        ```
        
        ## Usage
        
        For a complete introduction to `PFG`, and full coding examples, take a
         look at the sample notebook `example.ipynb`.
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Description-Content-Type: text/markdown
