Skip to content

clercrobin/HSMM_SimDyad

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

HSMM_SimDyad

============= HSMM based SimDyad is a python project using Bayesian Nonparametric Hidden semi-Markov Models to Disentangle Affect Processes During Marital Interaction This project is based on Matthew J. Johnson's work <pyhsmm>.

Note: HSMM_SimDyad is not based on a fork of pyhsmm. Instead, an old version of pyhsmm has been modified to achieve the specific tasks of HSMM SimDyad.

Authors:

William A. Griffin,Center for Social Dynamics and Complexity, Arizona State University

Xun Li, Arizona State University

Data

The data used for HSMM_SimDyad is located at:

.\matAll\

Each file is a sequential data for a specific couple (e.g. c200.txt represents data for c200).

Usage

Dependencies: wxpython, numpy, scipy.

To run it, just simple call

python hsmm.py

If you want a GUI, call

python hsmmSimDyad.py

If you want to run in parallel, call

python hsmmHPC.py

For the paper "Using Bayesian Nonparametric Hidden semi-Markov Models to Disentangle Affect Processes During Marital Interaction", we have setup 5 experiements for randomization testing -- check the shell script for details

python batchScriptRn1.sh
python batchScriptRn2.sh
python batchScriptRn3.sh
python batchScriptRn4.sh
python batchScriptRn5.sh

To configure the parameters or hyper-parameters, please ref

./conf/gen_params.py
./gcp/couple_gaussianprior.py

About

HSMM base SimDyad

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 38.6%
  • C++ 24.9%
  • Makefile 13.7%
  • CMake 10.0%
  • Fortran 7.0%
  • AGS Script 2.4%
  • Other 3.4%