search

mne-tools / mne-python

MNE: Magnetoencephalography (MEG) and Electroencephalography (EEG) in Python
https://mne.tools
BSD 3-Clause "New" or "Revised" License
2.67k stars 1.31k forks source link

mne does not return event labels using find_events or _read_vmrk_events #5721

Closed russellu closed 3 years ago

russellu commented 5 years ago

Hello, i am trying to load all the events for a given brainvision dataset (.vhdr, .eeg, .vmrk). however, i haven't been able to get the actual event labels using any of the functions i've found so far, and it drops any event that isn't prefixed by "Response"

my marker file (.vmrk) has a total of 5127 events, each with string labels, and latencies, here are examples of 3 types of events in my .vmrk: 1) Mk3382=Response,R1,36568476,1,0 2) Mk3395=Bad Interval,BadMin-Max,13132,187,0 3) Mk5011=Scoring,NREM2,960000,0,0

mne.find_events: returns a 3391 x 3 array, so it returns all the R1 markers (example #1) but ignores the Bad Interval and Scoring markers.

brainvision._read_vmrk_events: returns a 3630 x 3 array, it returns all the R1 and Scoring markers, but ignores the Bad Interval markers.

is there a way to simply load all the events, to get an array with the trigger label and latency? from there, i can find the markers i need.

russellu commented 5 years ago

i should note that some of these events were added in after the scan, as scoring evens (ie, the Bad Interval and Scoring events). not sure if that matters.

agramfort commented 5 years ago

what version of MNE are you using?

do you see them in raw.annotations provided you use current master?

russellu commented 5 years ago

raw.annotations is empty.

here is my mne system info:

Platform: Linux-3.16.0-4-amd64-x86_64-with-debian-8.0 Python: 3.6.6 | packaged by conda-forge | (default, Jul 26 2018, 09:53:17) [GCC 4.8.2 20140120 (Red Hat 4.8.2-15)] Executable: /home/brain/anaconda3/bin/python CPU: : 6 cores Memory: 26.6 GB

mne: 0.16.2 numpy: 1.15.3 {blas=mkl_rt, lapack=mkl_rt} scipy: 1.1.0 matplotlib: 3.0.0 {backend=module://ipykernel.pylab.backend_inline}

sklearn: 0.20.0 nibabel: 2.3.1 mayavi: Not found pycuda: Not found skcuda: Not found pandas: 0.23.4

agramfort commented 5 years ago

ok you need to try the current nightly version. We have significantly improved support of marker/annotations for BrainVision files.

can you try and tell us how it goes?

thanks

hoechenberger commented 3 years ago

This was probably fixed long ago.