What’s new?#

Here we list a changelog of MNE-connectivity.

Version 0.8 (in dev)#

Minimum supported Python version is now 3.10.

Add a new CoherencyDecomposition class for decomposing connectivity sources using multivariate coherency-based methods, by Thomas Binns (#193).
Add new plotting methods CoherencyDecomposition.plot_filters() and CoherencyDecomposition.plot_patterns() for visualising the decomposed connectivity sources, by Thomas Binns (#208 and #280).
Add support for computing multiple components of multivariate connectivity in the spectral_connectivity_epochs() and spectral_connectivity_time() functions and CoherencyDecomposition class, and add support for storing data with a components dimension in all Connectivity classes, by Thomas Binns and Eric Larson (#213).
Add support for mne.time_frequency.EpochsSpectrum objects to be passed as data to the spectral_connectivity_epochs() function, by Thomas Binns and Eric Larson (#220).
Add support for mne.time_frequency.EpochsTFR objects to be passed as data to the spectral_connectivity_epochs() and spectral_connectivity_time() functions, by Thomas Binns and Daniel McCloy (#232).
Update the cross-references for relevant functions and classes and make data types more explicit throughout the documentation, by Thomas Binns (#214).

Add a new min_distance parameter to the plot_sensors_connectivity() function which offers greater control over the minimum distance required between sensors to plot a connection between them, by Thomas Binns and Eric Larson (#221).
Begin deprecation of 2*np.pi as the default value of sfreq in favour of None in phase_slope_index(), by Thomas Binns (#326).