Skip to content
forked from guyz/HMM

Python Hidden Markov Models framework

Notifications You must be signed in to change notification settings

huazai-beyond121/HMM

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

HMM

A numpy/python-only Hidden Markov Models framework. No other dependencies are required.

This implementation (like many others) is based on the paper: "A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition, LR RABINER 1989"

Major supported features:

  • Discrete HMMs
  • Continuous HMMs - Gaussian Mixtures
  • Supports a variable number of features
  • Easily extendable with other types of probablistic models (simply override the PDF. Refer to 'GMHMM.py' for more information)
  • Non-linear weighing functions - can be useful when working with a time-series

Open concerns:

  • Examples are somewhat out-dated
  • Convergence isn't guaranteed when using certain weighing functions

About

Python Hidden Markov Models framework

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%