The official Wyrm repository has a new home at the BBCI github account. Please use the new one for forks, issues, etc.
Wyrm is a Brain Computer Interface (BCI) toolbox written in Python. Wyrm is suitable for running on-line BCI experiments as well as off-line analysis of EEG data.
Online documentation is available here.
Use distutils to install Wyrm into your PYTHONPATH:
$ git clone http://github.com/venthur/wyrm
$ cd wyrm
$ python setup.py install --userthis will always give you the latest development version of Wyrm.
Wyrm is also available on the Python Package Index (PyPI) and can be easily installed via:
$ pip install wyrmIn the examples directory, you'll find, among others, examples for various BCI
tasks using publicly available BCI datasets from the BCI Competition.
-
An example for classification of motor imagery in ECoG recordings. For that example the BCI Competition3, Data Set 1 was used.
-
An example for classification with a P300 Matrix Speller in EEG recordings. The BCI Competition 3, Data Set 2 was used for that example.
You can follow those examples by downloading the data and copying the files to the appropriate places.
Wyrm is mainly developed under Python 2.7, however since people will eventually move on to Python 3 we try to be forward compatible by writing the code in a way that it runs on Python 2 and -3.
Whenever a new version of Wyrm is pushed to github, the Travis continuous integration service will run Wyrm's whole test suite with Python 2.7, 3.3, and 3.4. If and only if all three test suites pass, the build is shown as "passing".
For a complete BCI system written in Python use Wyrm together with Mushu and Pyff. Mushu is a BCI signal acquisition and Pyff a BCI feedback and -stimulus framework.
If you use Wyrm for anything that results in a publication, We humbly ask you to cite us:
@Article{venthur2015,
author={Venthur, Bastian and Dähne, Sven and Höhne, Johannes and Heller, Hendrik and Blankertz, Benjamin},
title={Wyrm: A Brain-Computer Interface Toolbox in Python},
journal={Neuroinformatics},
year={2015},
volume={13},
number={4},
pages={471--486},
issn={1559-0089},
doi={10.1007/s12021-015-9271-8},
url={http://dx.doi.org/10.1007/s12021-015-9271-8}
}