v1.9.0

What's Changed

👁️ User-visible changes

• Add number of subjects to grand-average report by @SophieHerbst in #902, #910
• Drop Optional and Union from type hints by @hoechenberger in #908
• Drop support for loose=None by @hoechenberger in #915
• Streamline ICA reporting by @larsoner in #899
• Support variable time windows for epochs metadata creation by @hoechenberger in #874
• Raise a more helpful error message when requested subjects are missing from the dataset by @hoechenberger in #928
• In the report TOC, don't put the run numbers in quotation marks by @hoechenberger in #933
• Rename ica_ctps_ecg_threshold to ica_ecg_threshold by @hoechenberger in #935
• Improve logging output when rejecting epochs; and log number of reconstructed ICA and SSP epochs by @hoechenberger in #936
• Improve logging during ICA artifact discovery and rejection by @hoechenberger in #937
• Save ECG and EOG evokeds created during ICA artifact detection to disk by @hoechenberger in #938
• Provide a cleaner traceback on error by @larsoner in #951
• Rework how we apply EEG montages by @hoechenberger in #960

📦 Dependencies

• Drop support for Python 3.9 by @hoechenberger in #908

🐛 Bug fixes

• Change default for info to use for inverse by @SophieHerbst in #919
• Fix report section on original raw data (didn't contain bad channels and subject or experimenter name) by @hoechenberger in #931

📚 Documentation

• Restructure configuration options documentation sections by @hoechenberger in #914
• Enhance documentation of caching by @SophieHerbst in #918
• Improve documentation and config validation of loose and depth parameters by @hoechenberger in #915
• Render type annotations in the documentation again by @hoechenberger in #909
• Do not show Annotated types in configuration options documentation by @hoechenberger in #911, #913
  *Fix typo by @larsoner in #940

⛑️ Code health

• Ensure input changes cause output changes by @larsoner in #904
• Cleaner hook into mkdocstrings-python by @larsoner in #943
• Remove ref-names from .git_archival.txt by @sappelhoff in #948
• Switch to trusted publishing by @tsbinns in #957
• Set up Dependabot to automatically keep GitHub Actions up-to-date by @hoechenberger in #893

🤖 Automated infrastructure updates

• [dependabot]: Bump actions/setup-python from 4 to 5 by @dependabot in #894
• [dependabot]: Bump actions/download-artifact from 3 to 4 by @dependabot in #895
• [dependabot]: Bump actions/upload-artifact from 3 to 4 by @dependabot in #896
• [dependabot]: Bump codecov/codecov-action from 3 to 4 by @dependabot in #898
• [dependabot]: Bump actions/checkout from 3 to 4 by @dependabot in #897
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #903
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #917
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #923
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #929
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #942
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #945
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #947
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #953
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #954
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #955
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #959

👩‍💻New Contributors

• @dependabot made their first contribution in #894
• @tsbinns made their first contribution in #957

Full Changelog: v1.8.0...v1.9.0

v1.8.0

What's Changed

👁️ User-visible changes

• We fixed a crash that was affecting users running CSP on Maxwell-filtered data by @larsoner in #890
• Print longer tracebacks in case of a crash while saving a report file by @larsoner in #892

📦 Dependencies

• Explicitly depend on annotated-types by @hoechenberger in #886

📚 Documentation

• Add missing release dates to changelog by @hoechenberger in #887

⛑️ Code health

• We removed the unused settings shortest_event and study_name. They were relics of the early days of the pipeline and haven't been in use for a long time by @hoechenberger and @larsoner in #888, #889
• We now internally check that all generated configuration options are actually used somewhere by @larsoner in #889

🤖 Automated infrastructure updates

• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #891

Full Changelog: v1.7.0...v1.8.0

v1.7.0

👁️ User-visible changes

• Improve logging during cache invalidation by @hoechenberger in #877
• Fix handling of analyze_channels for EEG data, where some steps (like ERP calculation) would previously fail by @hoechenberger in #883

📚 Documentation

• Update contributing instructions and remove outdated building instructions by @hoechenberger in #875
• Fix a typo in the readme by @hoechenberger in #871

⛑️ Code health

• Enable stricter linting by @hoechenberger in #872

🤖 Automated infrastructure updates

• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #885

Full Changelog: v1.6.0...v1.7.0

v1.6.0

What's Changed

• The "Behavior changes" heading was incorrectly commented out in the changelog by @hoechenberger in #823
• MAINT: Fix CIs by @larsoner in #824
• Switch build backend to hatchling by @hoechenberger in #825
• FIX: Paths and pyarrow by @larsoner in #834
• MAINT: Test result caching by @larsoner in #836
• MAINT: Enable more ruff rules by @larsoner in #838
• ENH: Add artifact regression by @larsoner in #837
• BUG: Fix several bugs by @larsoner in #839
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #843
• ENH: Write out raw data and SSP events by @larsoner in #840
• BUG: Fix bug with Maxwell step when find_noise_channels_meg=False by @larsoner in #847
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #851
• BUG: Fix bug with multichannel classification by @larsoner in #853
• ENH: Better event count rep and noise_cov_method by @larsoner in #854
• BUG: Fix bug with split epochs and ICA by @larsoner in #855
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci in #856
• DOC: Automatic flowchart by @larsoner in #860
• DOC: Automatically document settings by @larsoner in #863
• add image_kwargs to report by @SophieHerbst in #859
• MAINT: Remove datalad by @larsoner in #867
• ENH: Split ICA into multiple steps by @larsoner in #865
• ENH: Document relevant steps by @larsoner in #866
• ENH: Doc config in sheet by @larsoner in #869

Full Changelog: v1.5.0...v1.6.0

v1.5.0 - Improved caching, movement compensation, and decoding

What's Changed

• MAINT: Cleanup dependencies and doc build by @larsoner in #755
• ENH: Add annotation of head velocity by @larsoner in #757
• BUG: Fix bug with cache invalidation by @larsoner in #756
• ENH: Add T1 and FLASH BEM to website by @larsoner in #758
• BUG: Fix bug with empty room processing by @larsoner in #761
• ENH: More compact logging by @larsoner in #764
• ENH: Add eSSS by @larsoner in #762
• ENH: Cache grand-average steps by @larsoner in #765
• MAINT: Test MD5 hashing too by @larsoner in #768
• ENH: Better logging of distances by @larsoner in #769
• MAINT: pre-commit.ci by @larsoner in #774
• Added deriv_root argument to CLI by @vferat in #773
• MAINT: Use pooch for downloading web data by @larsoner in #775
• API: Change default to _cache by @larsoner in #778
• ENH: Validate inputs with pydantic by @larsoner in #779
• TST: Rerun CIs by @larsoner in #781
• BUG: Fix bug where individual invs were not rendered by @larsoner in #782
• MAINT: Fix bug with item by @larsoner in #791
• Ensure link in "You're not viewing the latest stable version" banner is readable in light and dark theme by @hoechenberger in #793
• Reduce logging output when creating and saving reports by @hoechenberger in #799
• Use inst.pick() instead of inst.pick_types() by @hoechenberger in #800
• Use cleaned epochs for sensor-space decoding, time-frequency analysis, and covariance calculation by @hoechenberger in #796
• Add "picard-extended_infomax" ICA algorithm by @hoechenberger in #801
• Do not reject epochs based on PTP threshold before ICA cleaning by @hoechenberger in #806
• Add "autoreject_local" as new option for preprocessing/ptp_reject by @hoechenberger in #807
• ENH: Table sorting and ds004229 updates by @larsoner in #809
• Add support for autoreject (local) before running ICA by @hoechenberger in #810
• BUG: Fix bug with verbosity by @larsoner in #813
• Do not interpolate bad segments using autoreject before ICA by @hoechenberger in #816
• Omit bad channels when gathering data used for decoding by @hoechenberger in #817
• Allow decoding using epochs at different stages of pre-processing by @hoechenberger in #819
• Prepare for 1.5.0 release by @larsoner in #821
• FIX: PyQt6 by @larsoner in #822

New Contributors

• @vferat made their first contribution in #773

Full Changelog: v1.4.0...v1.5.0

v1.4.0 – Support for movement compensation

What's Changed

🆕 New features

• ENH: Add movement compensation by @larsoner in #747

🐛 Bug fixes

• BUG: Fix bug with mf_reference_run by @larsoner in #742
• Fix bug with too many JSON files found during empty room matching by @allermat in #743
• Fix documentation examples for ch_types configuration option by @allermat in #745
• Fix for documentation issue related to processing EEG separately in a combined M/EEG dataset by @allermat in #751

⚕ Code health & maintenance

• MAINT: Refactor _get_raw_paths by @larsoner in #749
• MAINT: DRY kwarg passing by @larsoner in #746
• MAINT: Use a large resource class to speed up build by @larsoner in #741
• Prepare for 1.4.0 release by @hoechenberger in #753

New Contributors

• @allermat made their first contribution in #743

Full Changelog: v1.3.0...v1.4.0

v1.3.0 - Bug fixes

What's Changed

• Update documentation, bump MNE-Python dependency, and prepare 1.3 rel… by @hoechenberger in #727
• BUG: Fix bug with sys info addition by @larsoner in #732
• Re-enable the copy button for code blocks in the documentation by @hoechenberger in #735
• Provide a more helpful log message if the CSP decoding step is being skipped by @hoechenberger in #734
• Always generate grand-averaged CSP decoding results if CSP was used on single-subject level by @hoechenberger in #733
• MAINT: Dont use deprecated book API by @larsoner in #736
• MAINT: Use rich instead of coloredlogs by @larsoner in #737
• MAINT: Deal with h5py numpy compat by @larsoner in #738

Full Changelog: v1.2.0...v1.3.0

v1.2.0 – Reduce start-up and processing time

User-facing changes

• Use caching in more places to reduce start-up and processing time, esp. when processing many subjects, by @agramfort in #716 and by @apmellot in #720
• Some documentation improvements by @hoechenberger in #712

Code health and maintenance

• Replace plot_psd() with compute_psd().plot() by @drammock in #725
• Bump required mkdocs-material version to ensure Twemojis work by @hoechenberger in #708
• Use git by @larsoner in #719
• Fix bug with dev version numbering by @larsoner in #718
• Start 1.2 release cycle by @hoechenberger in #707

Full Changelog: v1.1.0...v1.2.0

v1.1.0

What's Changed

User-facing changes

• Add notch filter by @hoechenberger and @larsoner in #702

Code health and maintenance

• Sync readme contents with built documentation by @hoechenberger in #698
• FIX: Dedup by @larsoner in #699
• MAINT: Rotate CircleCI key by @larsoner in #700
• Prepare for 1.1 release by @hoechenberger in #703

Full Changelog: v1.0.3...v1.1.0

v1.0.3

What's Changed

• BUG: Fix bug with tag deployment by @larsoner in #683
• MAINT: Fix bug with overrides by @larsoner in #685
• MAINT: Ignore, CIs, pre-commit hook by @larsoner in #689
• STY: Reformat with black by @larsoner in #690
• MAINT: Black infra by @larsoner in #691
• FIX: git setup by @larsoner in #692
• MAINT: Fix N_JOBS to n_jobs by @larsoner in #696
• Various doc improvements by @hoechenberger in #693
• DOC: Document as 1.0.3 by @larsoner in #697

Full Changelog: v1.0.2...v1.0.3