[Mne_analysis] Epoch() classifies all epochs as "bad"

Daniel Preciado daniel.preciadov at gmail.com
Tue May 29 11:34:40 EDT 2018
Search archives:

        External Email - Use Caution        

Dear list,

I am having trouble with epoching eeg data. For some reason, the epoch()
call classifies all epochs as bad, and I always end up with an empty Epochs
object.

If I try epoching with "preload=False", it reads correctly the epochs, and
the detected epochs perfectly matching the amount of trials and other
experiment parameters. But whenever I try to do anything with the Epochs
object (e.g. plotting), all events are deleted, and thus everything else
throws zeroDivisionErrors. And if I call back the epochs object after this,
it is empty (epoch counts are set to 0).

I have tried all types of combinations of arguments, but the result is
always the same: It seems to correctly recognize the epochs if the data is
not Preloaded, but everything is discarded if the data is preloade, or if I
try to do anything with the epochs object. I have tried different things
with the arguments, including reject=None, Baseline=(None, None),
preload=False, reject=None, flat=None, reject_tmin=None, reject_tmax=None,
detrend=None, on_missing='error', metadata=None, verbose=None. I have also
tried creating the epochs with the minimum arguments necessary (Data,
Event_data, Event_id), but it is always the same result.

I have also tried the same code on different datasets, but it always
discards all epochs, no matter what.

I do not understand why Epoch() classifies all data (from different people,
different sessions) as bad, can anyone help me to figure this out, and
hopefully fix it?

Thanks!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20180529/a9e030f8/attachment.html 


More information about the Mne_analysis mailing list