search

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
133 stars 65 forks source link

use tsss with mf_st_duration = signal duration #952

Open SophieHerbst opened 1 month ago

SophieHerbst commented 1 month ago

We have a dataset with specific artifacts, and are trying to apply tsss with mf_st_duration set to the signal duration. The problem is that the resting state run is much shorter than the experimental run, and we cannot set one duration that fits all signals. What would be the best fix for this? Can we set two versions of the mf_st_duration per task? Thanks in advance!

larsoner commented 1 month ago

One option would be to internally use min(st_duration, raw.times[-1]) and document that behavior. Then in your code you could set st_duration = 10000 or whatever and it would automatically be trimmed to the duration of each run.

SophieHerbst commented 3 weeks ago

For now we decided to use the duration of the resting state run. Would it be worth making the use of raw.times[-1] a config option?

larsoner commented 3 weeks ago

Rather than explicitly making that an option I would follow my comment one above ☝️ where you can just set st_duration=10000 or whatever