BUG: Fix bug with empty room processing

mne-tools / mne-bids-pipeline

Automatically process entire electrophysiological datasets using MNE-Python.

https://mne.tools/mne-bids-pipeline/

BSD 3-Clause "New" or "Revised" License

140 stars 67 forks source link

BUG: Fix bug with empty room processing #761

Closed larsoner closed 1 year ago

larsoner commented 1 year ago

Before merging …

[x] Changelog has been updated (docs/source/changes.md)

@allermat can you try this branch? In theory you should not need to delete any derivatives, just update to this branch and run.

Closes #759

Only mentioning the epochs bug fix here since the one @allermat is actually hitting is a within-release (within the last week I think!) bugfix

larsoner commented 1 year ago

Okay turns out the epochs were okay because we already use the min across runs, so I adjusted the change log and title. I still think #760 is probably the best way to deal with this issue for now, especially when (later) combined with empirical rank estimation which will be necessary for movement compensation.

larsoner commented 1 year ago

@hoechenberger feel free to merge if you're happy, I think it should fix the bug but even if it doesn't is a nice/helpful refactor (again)

larsoner commented 1 year ago

... I also realized that the head pos text file was being written as ds004229/sub-102/meg/sub-102_task-amnoise for example, this fixes it so it's written as ds004229/sub-102/meg/sub-102_task-amnoise_headpos.txt

hoechenberger commented 1 year ago

Thanks @larsoner