[Eeglablist] re : proper importing of events into eeglab

[Tarik S Bel-Bahar]

Hello Vadlamani Samhitha, some notes below. best wishes.


0. If you haven't had a chance to yet, be sure to Google eeglablist + your
topic (events or changing events) to review past responses about your
topic, from this year and past years. Also, See the eeglab tutorial and
google past eeglablist posts on importing events, changing events from the
gui, and changing events with a script

1. channel import should not be messing with your event import. Try to have
all your channels, events, etc... correct in a eeglab format before moving
to bcilab.

2. you may need to modify the events list (within eeglab) to make the
labels shorter "Target1" and "Target2". . There are several messages like
this about "Events" in the recent eeglablist.
You could have a script that runs through all the EEG.events, and whenever
it see Targetcode1 it changes it to Target1, etc...
See also the eeg_addnewevents function mentioned on the eeglablist, and
also googlable. THis would allow you to make a long list of labels and
times, and then apply them to a file all at once.

3. Note there are also options to import a "text" list of events (import
events), and to "assign new events" or "modify old ones" with a matlab
script.

4. if you add or modify an event from the gui, then type eegh to see the
code that was run, and then use that code in a matlab script to modify all
your events.

5. If you have specific questions about BCILab, be sure to also contact the
developer, C.Kothe, check out the online video tutorials for bcilab too if
you haven't had a chance to. My understanding is that the bcilab functions
are robust and well-documented, so you should be able to
open/review-documentation further for any bcilab function that you're using.













On Mon, Sep 19, 2016 at 12:31 PM, Vadlamani Samhitha <
vadlamani.samhitha at gmail.com> wrote:

> Hi . I am using the dataset for motor imagery from physionet.org . I need
> to process the data in BCILAB.Since , I couldn't import channel location
> file in bcilab explicitly , I was suggested to import the channels using
> eeglab and use the resulting .set file in bcilab . However after doing so ,
> two events in bcilab namely , 'Targetcode1' and 'Targetcode2' have
> converged into a single event namely , 'Targetcode' when it comes to eeglab
> . So when I import back this file with channels into bcilab , during
> training , an error pops up saying " your training data set has no trails
> for one of your classes ; you need at least two classes to train a
> classifier " which I'm assuming is because the Targetcode1 and Targetcode2
> in the raw data set have become Targetcode( a single class) in eeglab .
> Could someone please tell me how I can ensure that the events have been
> imported properly into eeglab. Thank you.
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.
> ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20160919/494ea6c6/attachment.html>