[Eeglablist] Error while rejecting continuous data (eeglab 14)

Sat Apr 22 07:17:31 PDT 2017

Hello Rafal,
I used your code and I noticed that all events were kept after cleaning. However, it seems it is not possible to convert the events back to numbers after cleaning with the reverse code of EEG.event(1).type = str2num(EEG.event('1').type). Do you have any idea what to do? I need them back to normal after cleaning. 
I emailed you before but no response. I am a new user of MATLAB and other related software programs. I wonder if you could help me out please.
Best regards,Thank you in advance for your help 

    On Monday, 17 April 2017, 13:45, Rafał Jończyk <rafal.jonczyk at gmail.com> wrote:

 Dear all,
I have found a simple solution, but I'd like to get some feedback if this does not negatively impact the EEG.event.structure in any way.What I did was to write the following code:
EEG.event(1).type = num2str(EEG.event(1).type)
, which changes the first event in EEG.event.type structure field into a string. Next, eeglab seems to automatically convert all the other events into strings to make it a homogeneous array.
Would you say that is okay? I already checked, and I can insert boundary events without any trouble.
Best,Rafał
2017-04-15 11:55 GMT+02:00 Rafał Jończyk <rafal.jonczyk at gmail.com>:

Dear Nicholas and Zahra,
thank you for your feedback. I thought that the problem might lie in the numeric/string nature of events. I am a bit confused, though, because eeglab gave me a warning message a few times saying "converting all event types to strings" -- could it be that the conversion didn't work? Also do you know if there is a way to run such a conversion manually without doing any harm to the event structure? In other words -- is there an unambiguous and safe way to convert the numeric event types to string event types?
Best,Rafal

2017-04-13 14:31 GMT+02:00 zahra fotovatnia <fotovatnia at yahoo.com>:

Hi Rafal,
Yes. I found the same problem when using cleaning continuous data. The point is I was trying to do difference waves on erplab after analyzing continuous data on eeglab but I repeatedly found the error of missing data. I tried different options but the same error. ASR deleted the events while cleaning.
BestZahra

    On Thursday, 13 April 2017, 2:45, Rafał Jończyk <rafal.jonczyk at gmail.com> wrote:

 Thank you Tarik for advice. I will have a go at your suggestions. I think there might be something with the latest version of eeglab given that I have never experienced such problems in previous versions, and it's not the first time that I am analyzing *cnt files from neuroscan in eeglab.
Has anyone else experienced a similar problem?
Best,Rafał
2017-04-12 3:10 GMT+02:00 Tarik S Bel-Bahar <tarikbelbahar at gmail.com>:

Hello Rafal, a few notes below that may help, all the best!
****************************** **Yes you could try just changing the format of all the events by looping through them and rewriting them.It may not be about the numeric format, it may be how they are read and/or stored (cell vs. strong array, or something basic like that).See also addnewevents function mentioned on list previously, and some related messages,which may allow you to write in new events that are accepted by eeglab.However, this may be problem with new build as you're seeing, thus a bugzilla note would be cool when you have a chance.Another thing to check is if the same thing happens with the eeglab tutorial data too,which would be telling if there's something unique about your data and/or it's event structure.

On Tue, Apr 11, 2017 at 10:37 AM, Rafał Jończyk <rafal.jonczyk at gmail.com> wrote:

Dear eeglablist community,
I have experienced some problems with event structure -- that's my guess -- whenrejecting fragments of continuous data (raw *.cnt files resampled to 250Hz& filtered) in the latest version of eeglab. Interestingly, this does not happen when I amusing eeglab 13.6.5b on the same computer. Note that I am not rejecting the onset of the data.
The error:

eeg_insertbound(): 3 boundary (break) events added.
eeg_insertbound(): 3 boundary (break) events added.
Warning: converting all event types to strings
BUG 1971 WARNING: IF YOU ARE USING A SCRIPT WITTEN FORA PREVIOUS VERSION OF
EEGLAB TO CALL THIS FUNCTION, BECAUSE YOU AREREJECTING THE ONSET OF THE DATA,
EVENTS WERE CORRUPTED. EVENT LATENCIES ARE NOW CORRECT(SEE https://sccn.ucsd.edu/wiki/EEG LAB_bug1971);

Error using cell/strmatch (line 19)
Requires character array or cell array of strings asinputs.

Error in eeg_eegrej (line 147)
    indBound2 =strmatch('boundary', { event2(:).type });
 Error while evaluating DestroyedObject Callback

 Could you give me a hint how to resolve the problem and continue working with the latest version of eeglab? Is it because my events are numeric? Best,
Rafał
-- 
Rafał Jończyk
Assistant Professor
Faculty of EnglishAdam Mickiewicz University, Poznań | Poland

______________________________ _________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/ee glabmail.html
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.uc sd.edu
For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.e du

-- 
Rafał Jończyk
Assistant Professor
Faculty of EnglishAdam Mickiewicz University, Poznań | Poland
______________________________ _________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/ee glabmail.html
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.uc sd.edu
For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.e du

-- 
Rafał Jończyk
Assistant Professor
Faculty of EnglishAdam Mickiewicz University, Poznań | Poland

-- 
Rafał Jończyk
Assistant Professor
Faculty of EnglishAdam Mickiewicz University, Poznań | Poland

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20170422/cf583195/attachment.html>