Skip to content

VandroiyLabs/chapolins

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 

Repository files navigation

Chapolins

A change point problems can be stated as follows: given a time series, how likely was there a change of distribution in those points? If there was, when that happened? Answering both questions is the main point of change point detection & analysis. Chapolins is a library we have been developing for nonparametric change point analysis, versatile and easily expanded. We designed it using scipy's look & feel.

By integrating C libaries, fast searching algorithms and some parallelism (being developed), we have been applying this especially into neuroscience, animal communication and electronic nose applications.

If you are using this package, please make sure to cite us Mosqueiro et al @ CISS 2016 and the authors of the methods we implemented (for an incomplete, but informative, list check our methods section below).

The code will be fully available in a few weeks, when the paper is published

How it works

Methods

The name

Chapolin stands for CHAnge POint LIbrary for Nonparametric Statistics, and is a small tribute to a group of comedians part of a series of TV Shows, one of which called El Chapulin Colorado (or Chapolin in brazilian portuguese). This was, and remains, a huge success in many countries.

Dependencies

Most of the library is written to interface from Python.

License

This is under Apache license, which can be found here.

About

Change Point Library for Nonparametric Statistics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published