jajapy 0.1.1

Baum-Welch for all kind of Markov model

jajapy

Introduction

jajapy is a python library implementing the Baum-Welch algorithm on various kinds of Markov models. Warning jajapy is still a WIP.

Main features

jajapy provides:

• BW algorithm for Hidden Markov Models reference
• BW algorithm for Markov Chains
• BW algorithm for Gaussian Observation Hidden Markov Models reference
• BW algorithm for Markov Decision Processes reference
• Active BW algorithm for Markov Decision Processes reference
• BW algorithm for CTMC
• BW algorithm for asynchronous parallel composition of CTMCs

Additionally, it provides other learning algorithms:

• Alergia, for Markov Chains reference
• IOAlergia, for Markov Decision Processes reference

Installation

pip install jajapy

Requirements

• numpy
• scipy

Get Started

Coming soon. For now check demo.py.

TO DO

• unit tests
• generate the documentation. Add examples.
• upload it to Pypi
• link with sotrmpy, prism
• error management