BUG: Fix bug with least-squares sphere fit #13178
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
larsoner
requested review from
agramfort,
alexrockhill,
dengemann and
drammock
as code owners
drammock
approved these changes
Mar 31, 2025
Member
drammock
left a comment
drammock
left a comment
There was a problem hiding this comment.
couple minor nits, otherwise LGTM. nice improvement!
larsoner
added a commit
to larsoner/mne-python
that referenced
this pull request
Apr 4, 2025
* upstream/main: (149 commits) FIX make_watershed_bem to handle missing talairach_with_skull.lta courtesy Freesurfer 8 (mne-tools#13172) ENH: Add upsampling for MEG helmet surface (mne-tools#13179) MAINT: Update code credit (mne-tools#13180) BUG: Fix bug with least-squares sphere fit (mne-tools#13178) fix EDF export (mne-tools#13174) fix typo (mne-tools#13171) [pre-commit.ci] pre-commit autoupdate (mne-tools#13164) Fix dev installation guide (mne-tools#13163) expose 'mode' for plotting dipole on brain (mne-tools#13162) turn dipole attrs into properties (mne-tools#13153) remove misformatted (and unused) crossref anchor (mne-tools#13155) doc: point to read_dipole (mne-tools#13149) [pre-commit.ci] pre-commit autoupdate (mne-tools#13152) BUG: Fix bug with not short-circuiting n_jobs=1 (mne-tools#13147) FIX: Missing coordinates.xml in MFF file (mne-tools#13148) FIX: Gracefully handle bad XML files in EGI reader (mne-tools#13145) Fixes for Latest IPython (9.0.1) (mne-tools#13146) Fix intersphinx (mne-tools#13143) BUG: Fix bug with parallel doc build (mne-tools#13140) [pre-commit.ci] pre-commit autoupdate (mne-tools#13141) ...
larsoner
added a commit
to SYXiao2002/mne-python
that referenced
this pull request
Apr 18, 2025
* upstream/main: (40 commits) fix typo (missing space) that messed up rst rendering (mne-tools#13217) MAINT: Restore VTK dev (mne-tools#13214) [pre-commit.ci] pre-commit autoupdate (mne-tools#13212) BUG: Fix bug with example (mne-tools#13210) MAINT: Fix pip-pre with PyVista (mne-tools#13207) Move FCBG to former partners (mne-tools#13205) ENH: Update related software list (mne-tools#13202) fix sfreq estimation for snirf files (mne-tools#13184) ENH: Use data-based padding instead of "odd" padding when filtering in raw.plot (mne-tools#13183) FIX: Bumps (mne-tools#13198) DOC: fix typo in examples/io/read_impedances.py (mne-tools#13197) [pre-commit.ci] pre-commit autoupdate (mne-tools#13173) FIX make_watershed_bem to handle missing talairach_with_skull.lta courtesy Freesurfer 8 (mne-tools#13172) ENH: Add upsampling for MEG helmet surface (mne-tools#13179) MAINT: Update code credit (mne-tools#13180) BUG: Fix bug with least-squares sphere fit (mne-tools#13178) fix EDF export (mne-tools#13174) fix typo (mne-tools#13171) [pre-commit.ci] pre-commit autoupdate (mne-tools#13164) Fix dev installation guide (mne-tools#13163) ...
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
SciPy recently reworked how
fmin_cobylaworks to use a more modern, robust library. This changed our fitting results a bit, which @drammock just tweaked some tolerances for.But it also made me look at our
_fit_spherecode, which is used primarily for theorigin="auto"head-origin fit inmaxwell_filter. It turns out this was never guaranteed to find the optimal (least-squares) sphere fit. So this updates our code to use a faster, more robust linear least-squares solution.This then allows us to remove a couple of
pytest.warnsthat had to do with bad sphere fits, because the fits are no longer bad (hooray!). I think for most applications this won't change much, but when there are some really bad outliers, this code should be more robust already because it doesn't rely on an initial center guess, which our old code did (and would be affected quite a bit by any outliers).