Mne

- Fix plotting of TFRs with linear y-scale (misaligned frequency bins), by Stefan Repplinger and Eric Larson
- Fix bug in `mne.preprocessing.ICA` where requesting extended infomax via `fit_params={'extended': True}` was overridden
- Fix handling of repeated events in `mne.Epochs`
- Fix reading of cardinals in .htps files (identifier are int not strings)
- Fix handling in cases where no components are found in `mne.preprocessing.corrmap`
- Fix bug in `mne.write_evokeds` where `evoked.nave` was not saved properly when multiple `mne.Evoked` instances were written to a single file
- Fix TAL channel parsing (annotations) for EDF-D files by
- Fix `info['sfreq']` when decimating in `mne.time_frequency.tfr_multitaper` and `mne.time_frequency.tfr_morlet` and make sure an error is raised when exceed Nyquist frequency
- Fix IO of TFRs when event_id contain a / in one of the keys

A few highlights
================
- Reorganized documentation, with 19 new or revised tutorials.
- Automatic MRI fiducial estimation based on MNI Talairach transforms.
- Improved plotting support, including new Butterfly plots for PSD epochs, and 3D sensor connectivity plots.
- Speed improvements in clustering, coregistration, volumetric source space creation, and other parts of the codebase.
- More supported file formats, including curry files, the updated NYU New York 2019 system for KIT, and the new Annotations support for CTF marker files.

In addition, we caught and fixed more than 41 bugs!

Notable API changes
===================
- New minimum supported dependencies, most critically **Python >= 3.5 is now required**.
- Complete reworking of EEG channel montage/digitization, including improvements for EEG source modeling with no MRI (surrogate MRIs).
- Fixes to volumetric morphing and plotting functions.
- Update of the FIF constants.

A few highlights
================

- Python 2 is no longer supported; MNE-Python now requires Python 3.5+
- New tutorials and examples on sleep stage classification, data simulation using subject's anatomy, how to use EEG montages on fsaverage
- New module to simulate SourceEstimates
- Improved performance using CUDA, better copy management, better dispatching of the computation over the channels
- New fetchers for polysomnography (PSG) recordings from Physionet and fsaverage template
- Better support for source reconstruction with beamformers and other inverse models
- Improved UI and visualizations for topomaps, Raw and Epochs objects
- Better support for Annotations
- Support to compute power envelope correlations on brain parcellation for rest data
- Better rendering of the coregistration
- Added partial support for PyVista as a 3D backend that can replace mayavi
- Added support for Raw, Epochs and Evoked noise simulation
- Added new parcellation (448-labels subdivided aparc) and improve support including morphing of the labels
- Better support for the TFR objects

And we caught and fixed more than 50 bugs!

Notable API changes
===================

- Deprecation of `mne.realtime` module to become entire project in the MNE echosystem
- Deprecation of `mne.io.find_edf_events`, `raw.estimate_rank`
- Reading BDF and GDF files with `mne.io.read_raw_edf` is deprecated and replaced by `mne.io.read_raw_bdf` and `mne.io.read_raw_gdf`
- The signatures of `mne.preprocessing.ICA`, `mne.simulation.add_noise` and `mne.simulation.add_chpi` have changed
- Added overwrite parameter in `mne.Epochs.save`
- `mne.minimum_norm.apply_inverse` now returns `mne.VolVectorSourceEstimate` when needed
- Annotations are now kept sorted by onset
- `peak_finder` should be imported as: `from mne.preprocessing import peak_finder`

A few highlights
================

- Python 2 is no longer supported; MNE-Python now requires Python 3.5+
- New tutorials and examples on sleep stage classification, data simulation using subject's anatomy, how to use EEG montages on fsaverage
- New module to simulate SourceEstimates
- Improved performance using CUDA, better copy management, better dispatching of the computation over the channels
- New fetchers for polysomnography (PSG) recordings from Physionet and fsaverage template
- Better support for source reconstruction with beamformers and other inverse models
- Improved UI and visualizations for topomaps, Raw and Epochs objects
- Better support for Annotations
- Support to compute power envelope correlations on brain parcellation for rest data
- Better rendering of the coregistration
- Added partial support for PyVista as a 3D backend that can replace mayavi
- Added support for Raw, Epochs and Evoked noise simulation
- Added new parcellation (448-labels subdivided aparc) and improve support including morphing of the labels
- Better support for the TFR objects

And we caught and fixed more than 50 bugs !!

Notable API changes
===================

- Deprecation of mne.io.find_edf_events, raw.estimate_rank
- Reading BDF and GDF files with mne.io.read_raw_edf is deprecated and replaced by mne.io.read_raw_bdf and mne.io.read_raw_gdf
- The signatures of mne.preprocessing.ICA, mne.simulation.add_noise and mne.simulation.add_chpi have changed
- Added overwrite parameter in mne.Epochs.save
- mne.minimum_norm.apply_inverse now returns mne.VolVectorSourceEstimate when needed
- Annotations are now kept sorted by onset
- peak_finder should be imported as: 'from mne.preprocessing import peak_finder'

Highlights
- This release will be the last release compatible with Python 2. The next version will be Python 3.5+ only.
- Better support for Annotations, including readers for EEGLAB, BrainVision, EDF, CSV, TXT and Brainstorm formats, and a new tutorial in the documentation to dive in.
- Better support to import data from FieldTrip and Neurmag 122 systems.
- Add capability to read and save Epochs containing complex data (e.g. after Hilbert transform).
- Add optically pumped magnetometer dataset and examples.
- New source morph object to unify morphing any type of source estimates (surface or volume) from one subject to another for group studies. It is now possible to do group studies when working on the volume with MNE.
- Add ability to read and write beamformers.
- New source power spectral estimation example for resting-state data
- Better support for Reports, now they can be load/saved in HDF5 and the existing figures from a report can be removed.
- Add support for reading MATLAB v7.3+ for EEGLAB.
- Add interactive visualization of volume source estimates using plot_volume_source_estimates
- New BIDS-compatible raw filename construction
- Better helmet visualization for Artemis123 and CTF

Notable API changes
- Deprecation of annot and annotmap parameters in mne.io.read_raw_edf
- Deprecated mne.SourceEstimate.morph_precomputed, mne.SourceEstimate.morph, mne.compute_morph_matrix, mne.morph_data_precomputed and mne.morph_data in favor of mne.compute_source_morph
- Calling mne.Epochs.decimate no longer copies the data when decim=1.
- Warning messages are now only logged when warn_explicit is set (unless a logging file is being used) to avoid duplicate warning messages.
- src.kind now equals to 'mixed' (and not 'combined') for a mixed source space (i.e., made of surfaces and volume grids)
- The default value of stop_receive_thread in mne.realtime.RtEpochs.stop has been changed to True
- Using the mne.io.Raw.add_channels on an instance with memmapped data will now resize the memmap file to append the new channels on Windows and Linux
- Mismatches in CTF compensation grade are now checked in inverse computation