Version 1.6.0 (2023-11-20)

Enhancements

• Add support for Neuralynx data files with mne.io.read_raw_neuralynx() (#11969 by new contributor Kristijan Armeni and new contributor Ivan Skelin)

• Improve tests for saving splits with mne.Epochs (#11884 by Dmitrii Altukhov)

• Added functionality for linking interactive figures together, such that changing one figure will affect another, see Using the event system to link figures and mne.viz.ui_events. Current figures implementing UI events are mne.viz.plot_topomap() and mne.viz.plot_source_estimates() (#11685 #11891 by Marijn van Vliet)

• HTML anchors for mne.Report now reflect the section-title of the report items rather than using a global incrementor global-N (#11890 by Eric Larson)

• Added public mne.io.write_info() to complement mne.io.read_info() (#11918 by Eric Larson)

• Added option remove_dc to to Raw.compute_psd(), Epochs.compute_psd(), and Evoked.compute_psd(), to allow skipping DC removal when computing Welch or multitaper spectra (#11769 by Nikolai Chapochnikov)

• Add the possibility to provide a float between 0 and 1 as n_grad, n_mag and n_eeg in compute_proj_raw, compute_proj_epochs and compute_proj_evoked to select the number of vectors based on the cumulative explained variance (#11919 by Mathieu Scheltienne)

• Add extracting all time courses in a label using mne.extract_label_time_course() without applying an aggregation function (like mean) (#12001 by Hamza Abdelhedi)

• Added support for Artinis fNIRS data files to mne.io.read_raw_snirf() (#11926 by Robert Luke)

• Add helpful error messages when using methods on empty mne.Epochs-objects (#11306 by Martin Schulz)

• Add support for passing a dict as sensor_color to specify per-channel-type colors in mne.viz.plot_alignment() (#12067 by Eric Larson)

• Add inferring EEGLAB files’ montage unit automatically based on estimated head radius using read_raw_eeglab(..., montage_units="auto") (#11925 by Jack Zhang, #11951 by Eric Larson)

• Add EpochsSpectrumArray and SpectrumArray to support creating power spectra from NumPy array data (#11803 by Alex Rockhill)

• Add support for writing forward solutions to HDF5 and convenience function mne.Forward.save() (#12036 by Eric Larson)

• Refactored internals of mne.read_annotations() (#11964 by Paul Roujansky)

• Add support for drawing MEG sensors in mne coreg (#12098 by Eric Larson)

• Improve string representation of mne.Covariance (#12181 by Eric Larson)

• Add check_version=True to mne sys_info to check for a new release on GitHub (#12146 by Eric Larson)

• Bad channels are now colored gray in addition to being dashed when spatial colors are used in mne.viz.plot_evoked() and related functions (#12142 by Eric Larson)

• By default MNE-Python creates matplotlib figures with layout='constrained' rather than the default layout='tight' (#12050, #12103 by Mathieu Scheltienne and Eric Larson)

• Enhance plot_evoked_field() with a GUI that has controls for time, colormap, and contour lines (#11942 by Marijn van Vliet)

• Add mne.viz.ui_events.UIEvent linking for interactive colorbars, allowing users to link figures and change the colormap and limits interactively. This supports plot_evoked_topomap(), plot_ica_components(), plot_tfr_topomap(), plot_projs_topomap(), plot_image(), and plot_image() (#12057 by Santeri Ruuskanen)

• Add example KIT phantom dataset in mne.datasets.phantom_kit.data_path() and KIT phantom dataset tutorial (#12105 by Judy D Zhu and Eric Larson)

• make_metadata() now accepts tmin=None and tmax=None, which will bound the time window used for metadata generation by event names (instead of a fixed time). That way, you can now for example generate metadata spanning from one cue or fixation cross to the next, even if trial durations vary throughout the recording (#12118 by Richard Höchenberger)

• Add support for passing multiple labels to mne.minimum_norm.source_induced_power() (#12026 by Erica Peterson, Eric Larson, and Daniel McCloy )

• Added documentation to mne.io.Raw.set_montage() and mne.add_reference_channels() to specify that montages should be set after adding reference channels (#12160 by Jacob Woessner)

• Add argument splash to the function using the qt browser backend to allow enabling/disabling the splash screen (#12185 by Mathieu Scheltienne)

• ICA’s HTML representation (displayed in Jupyter notebooks and mne.Report) now includes all optional fit parameters (e.g., max. number of iterations) (#12194, by Richard Höchenberger)

Bugs

• Fix bug where mne.io.read_raw_gdf() would fail due to improper usage of np.clip (#12168 by new contributor Rasmus Aagaard)

• Fix bugs with mne.preprocessing.realign_raw() where the start of other was incorrectly cropped; and onsets and durations in other.annotations were left unsynced with the resampled data (#11950 by new contributor Qian Chu)

• Fix bug where encoding argument was ignored when reading annotations from an EDF file (#11958 by new contributor Andrew Gilbert)

• Mark tests test_adjacency_matches_ft and test_fetch_uncompressed_file as network tests (#12041 by new contributor Maksym Balatsko)

• Fix bug with mne.channels.read_ch_adjacency() (#11608 by new contributor Ivan Zubarev)

• Fix bug where epochs.get_data(..., scalings=...) would errantly modify the preloaded data (#12121 by new contributor Pablo Mainar and Eric Larson)

• Fix bugs with saving splits for Epochs (#11876 by Dmitrii Altukhov)

• Fix bug with multi-plot 3D rendering where only one plot was updated (#11896 by Eric Larson)

• Fix bug where verbose level was not respected inside parallel jobs (#12154 by Eric Larson)

• Fix bug where subject birthdays were not correctly read by mne.io.read_raw_snirf() (#11912 by Eric Larson)

• Fix bug where warnings were emitted when computing spectra for channels marked as bad (#12186 by Eric Larson)

• Fix bug with mne.chpi.compute_head_pos() for CTF data where digitization points were modified in-place, producing an incorrect result during a save-load round-trip (#11934 by Eric Larson)

• Fix bug where non-compliant stimulus data streams were not ignored by mne.io.read_raw_snirf() (#11915 by Johann Benerradi)

• Fix bug with pca=False in mne.minimum_norm.compute_source_psd() (#11927 by Alex Gramfort)

• Fix bug with notebooks when using PyVista 0.42 by implementing trame backend support (#11956 by Eric Larson)

• Removed preload parameter from mne.io.read_raw_eyelink(), because data are always preloaded no matter what preload is set to (#11910 by Scott Huberty)

• Fix bug with mne.viz.Brain.get_view() where calling show_view() with returned parameters would change the view (#12000 by Eric Larson)

• Fix bug with mne.viz.Brain.show_view() where distance=None would change the view distance (#12000 by Eric Larson)

• Fix bug with add_annotation() when reading an annotation from a file with both hemispheres shown (#11946 by Marijn van Vliet)

• Fix bug with reported component number and errant reporting of PCA explained variance as ICA explained variance in mne.Report.add_ica() (#12155, #12167 by Eric Larson and Richard Höchenberger)

• Fix bug with axis clip box boundaries in mne.viz.plot_evoked_topo() and related functions (#11999 by Eric Larson)

• Fix bug with subject_info when loading data from and exporting to EDF file (#11952 by Paul Roujansky)

• Fix bug where mne.Info HTML representations listed all channel counts instead of good channel counts under the heading “Good channels” (#12145 by Eric Larson)

• Fix rendering glitches when plotting Neuromag/TRIUX sensors in mne.viz.plot_alignment() and related functions (#12098 by Eric Larson)

• Fix bug with delayed checking of info["bads"] (#12038 by Eric Larson)

• Fix bug with mne coreg where points inside the head surface were not shown (#12147, #12164 by Eric Larson)

• Fix bug with mne.viz.plot_alignment() where sensor_colors were not handled properly on a per-channel-type basis (#12067 by Eric Larson)

• Fix handling of channel information in annotations when loading data from and exporting to EDF file (#11960 #12017 #12044 by Paul Roujansky)

• Add missing overwrite and verbose parameters to Transform.save() (#12004 by Marijn van Vliet)

• Fix parsing of eye-link Annotations when apply_offsets=False is provided to read_raw_eyelink() (#12003 by Mathieu Scheltienne)

• Correctly prune channel-specific Annotations when creating Epochs without the channel(s) included in the channel specific annotations (#12010 by Mathieu Scheltienne)

• Fix plot_volume_source_estimates() with VolSourceEstimate which include a list of vertices (#12025 by Mathieu Scheltienne)

• Add support for non-ASCII characters in Annotations, Evoked comments, etc when saving to FIFF format (#12080 by Daniel McCloy)

• Correctly handle passing "eyegaze" or "pupil" to mne.io.Raw.pick() (#12019 by Scott Huberty)

• Fix bug with mne.time_frequency.Spectrum.plot() and related functions where bad channels were not marked (#12142 by Eric Larson)

• Fix bug with plot_raw() where changing MNE_BROWSER_BACKEND via set_config() would have no effect within a Python session (#12078 by Santeri Ruuskanen)

• Improve handling of method argument in the channel interpolation function to support str and raise helpful error messages (#12113 by Mathieu Scheltienne)

• Fix combination of DIN event channels into a single synthetic trigger channel STI 014 by the MFF reader of mne.io.read_raw_egi() (#12122 by Mathieu Scheltienne)

• Fix bug with mne.io.read_raw_eeglab() and mne.read_epochs_eeglab() where automatic fiducial detection would fail for certain files (#12165 by Clemens Brunner)

• Fix concatenation of raws with np.nan in the device to head transformation (#12198 by Mathieu Scheltienne)

• Fix bug with mne.viz.plot_compare_evokeds() where the title was not displayed when axes='topo' (#12192 by Jacob Woessner)

• Fix bug with mne.io.read_raw_cnt() where the bad channels were not properly read (#12189 by Jacob Woessner)

• Fix bug where iterating over Raw would result in an error (#12205 by Clemens Brunner)

API changes

• The default for mne.Epochs.get_data() of copy=False will change to copy=True in 1.7. Set it explicitly to avoid a warning (#12121 by new contributor Pablo Mainar and Eric Larson)

• mne.preprocessing.apply_maxfilter and mne maxfilter have been deprecated and will be removed in 1.7. Use mne.preprocessing.maxwell_filter() (see this tutorial) in Python or the command-line utility from MEGIN maxfilter and mne.bem.fit_sphere_to_headshape() instead (#11938 by Eric Larson)

• mne.io.kit.read_mrk() reading pickled files is deprecated using something like np.savetxt(fid, pts, delimiter="\t", newline="\n") to save your points instead (#11937 by Eric Larson)

• Replace legacy inst.pick_channels and inst.pick_types with inst.pick (where inst is an instance of Raw, Epochs, or Evoked) wherever possible (#11907 by Clemens Brunner)

• The reset_camera parameter has been removed in favor of distance="auto" in mne.viz.set_3d_view(), mne.viz.Brain.show_view(), and related functions (#12000 by Eric Larson)

• Several unused parameters from mne.gui.coregistration() are now deprecated: tabbed, split, scrollable, head_inside, guess_mri_subject, scale, and advanced_rendering. All arguments are also now keyword-only. (#12147 by Eric Larson)

Authors

• Adam Li

• Alex Rockhill

• Alexandre Gramfort

• Ana Radanovic

• Andy Gilbert+

• Clemens Brunner

• Daniel McCloy

• Denis A. Engemann

• Dimitri Papadopoulos Orfanos

• Dmitrii Altukhov

• Dominik Welke

• Eric Larson

• Erica Peterson

• Gonzalo Reina

• Hamza Abdelhedi

• Ivan Skelin+

• Ivan Zubarev+

• Jack Zhang

• Jacob Woessner

• Johann Benerradi

• John Veillette

• Judy D Zhu

• Kristijan Armeni+

• Mainak Jas

• Maksym Balatsko+

• Marijn van Vliet

• Martin Schulz

• Mathieu Scheltienne

• Nikolai Chapochnikov

• Pablo Mainar+

• Paul Roujansky

• Qian Chu+

• Rasmus Aagaard+

• Richard Höchenberger

• Rob Luke

• Santeri Ruuskanen

• Scott Huberty

• Stefan Appelhoff