rapfff/jajapy

Baum-Welch for all kind of Markov models

Introduction

jajapy is a python library implementing the Baum-Welch algorithm on various kinds of Markov models. jajapy generates models which are compatible with the Stormpy model checker. Thus, jajapycan be use as a learning extension to the Storm model checker.

Main features

jajapy provides:

Markov Model	Learning Algorithm(s)
MC	Baum-Welch for MCs Alergia (ref)
MDP	Baum-Welch for MDPs (ref) Active Baum-Welch (ref) IOAlergia (ref)
CTMC	Baum-Welch for CTMCs Baum-Welch for synchronous compositions of CTMCs
PCTMC	Baum-Welch for PCTMCs (ref)
HMM	Baum-Welch for HMMs (ref)
GoHMM	Baum-Welch for GoHMMs (ref)

jajapy is compatible with Prism and Storm.

Installation

pip install jajapy

Requirements

• numpy
• scipy
• alive-progress
• sympy
• stormpy (recommended: if stormpy is not installed, jajapy will generate models in jajapy format).

Documentation

Available on readthedoc

Reference and citation

• The extended version of the tool paper presented at QEST'23 is available here
• If you use this tool in your research, please cite it

About the author

My website

jajapy's Issues

No module named 'jajapy.hmm'

Hello Rapfff,

I am curious to use your tool jajapy. I installed it on my machine and I get the following error:

ModuleNotFoundError: No module named 'jajapy.hmm'

Do you have an idea of what could it be?

TypeError: MC_state() missing 1 required positional argument: 'nb_states'

Hello Rapfff,

I found another issue in the file demo.py at the moment of manually defining the Markov chain. The error is in the function modelMC_REBER line 5:

TypeError: MC_state() missing 1 required positional argument: 'nb_states'

From the description, nb_states: int number of states in which this state is

Let me know if you need further information.

RuntimeError

Thanks for updating the demo.py file! I am getting repeatedly the error below, is this normal? Maybe a simpler demo would work better?

    raise RuntimeError('''
RuntimeError: 
        An attempt has been made to start a new process before the
        current process has finished its bootstrapping phase.

        This probably means that you are not using fork to start your
        child processes and you have forgotten to use the proper idiom
        in the main module:

            if __name__ == '__main__':
                freeze_support()
                ...

        The "freeze_support()" line can be omitted if the program
        is not going to be frozen to produce an executable.
Fatal Python error: init_import_size: Failed to import the site module
Python runtime state: initialized

PS: If I may, it would be helpful to add a short description of the demo (e.g., a drawing of the markov model, the input parameters, outputs, and interpretation). I could provide this if I manage to run successfully the demo :-)