borsar 0.2

tools for electrophysiological analysis, especially cluster-based tests.

Various tools, objects and functions for EEG/MEG data analysis and visualisation. Some functionality that is available here may be later moved to mne-python.

borsar includes:

• PSD object for manipulation of power spectral results
• Clusters object for storage, manipulation and plotting of clutser-based results, both in channel and sourcee space
• efficient regression for multichannel data (compute_regression_t)
• cluster_based_regression to perform regression tests in cluster-based permutation framework
• numpy and numba implementations of cluster-based permutation tests in 3d space (for example in channels x frequency x time space) with optional filtering by minimum number of adjacent channels (min_adj_ch, equivalent of minnbchan in fieldtrip).
• Topo object for topomap plots that retains the topomap state, allows to mark channels, efficiently update data, change contour line width and style for one or multiple topomaps.

Installation

To get the official borsar version install using pip:

pip install borsar

The PyPI version may be sometimes behind the GitHub version so to install from GitHub you can use pip in the following way:

pip install git+https://github.com/mmagnuski/borsar

However, if you plan to frequently update the dev version and contribute to borsar, install by cloning the repo with git and installing in dev mode:

cd where_you_want_to_download_borsar
git clone https://github.com/mmagnuski/borsar
cd borsar
pip install -e .

the second and third method require git to be installed.

Documentation

Go to the online documentation for more information about usage examples or full API docs. :construction: be warned that documentation is under contstruction :construction: