[FieldTrip] Where to add a trialfun?
Ana Laura Diez Martini
diezmartini at gmail.com
Sun Oct 5 22:34:35 CEST 2014
Hi Jorn and all,
As I said before, everything works ok when I use ft_trialfun_general
instead of this trialfun. However, it is crucial for me to transform the
event values with that trialfun. And now I use 'trigger' instead of
'STATUS' but it doesn't change much. Any other idea? THANK YOU!
This is my script again:
% CONGRUENT TRIALS DEFINITION
cfg = [ ] ;
cfg.dataset = 'location';
event = ft_read_event('location');
cfg.trialdef.eventtype = 'trigger';
cfg.trialdef.eventvalue = cgrmrk;
cfg.trialdef.prestim = 0.2;
cfg.trialdef.poststim = 1;
cfg.trialfun = 'trialfun_bit2dec';
[cfg] = ft_definetrial(cfg);
On Wed, Oct 1, 2014 at 4:03 AM, "Jörn M. Horschig" <
jm.horschig at donders.ru.nl> wrote:
> Hi Ana,
>
> from the first call with eventtype = '?', the output says
>
>> event type: 'trigger' with event values: 'DIN1' 'DIN2' 'DIN4' 'DIN8'
>>
>
> This means that there is one event type in our data, namely 'trigger', and
> the only trigger values in your data are 'DIN1', 'DIN2', 'DIN4' and 'DIN8'.
> FieldTrip works that way that it passes on the cfg.trialdef structure to
> the trialfun that is specified. In your particular case, trialfun_bit2dec
> does not seem to have a very user-friendly error message, but the error
> message means that whatever eventtype and eventvalues you asked for, they
> are not present in your data. When looking at what you define, you can
> probably easily spot that you ask for cfg.trialdef.eventtype = 'STATUS'.
> However, checking back the output that you created with '?', there is no
> event type 'STATUS' in your data. Therefore, everything works fine with
> your trialfun. If you try specifying the trialdef properly, everything
> should work fine.
>
> A general advise when using an own trial fun and getting an error message:
> Try your code without specifying cfg.trialfun first. Then FieldTrip will
> use the standard trialfun, which should work if all your cfg-settings are
> fine. If the standard trialfun runs without errors, then your trialfun
> should do so too. If your trialfun, however, then crashes, the error is
> most likely in the trialfun that you used, and not in the cfg-settings. In
> your case, however, I am fairly certain that the cfg-settings you specified
> were causing your problem.
>
> I hope this helps.
> Best,
> Jörn
>
>
>
>
> On 9/30/2014 10:27 PM, Ana Laura Diez Martini wrote:
>
>> For a more simple check, when I try to read the events without the
>> trialfun ( so it uses ft_trialfun_general by default) I get the original
>> event values:
>>
>> >> cfg = [];
>> >> cfg.dataset = '27CW1.RAW';
>> >> cfg.trialdef.eventtype = '?';
>> >> cfg=ft_definetrial(cfg);
>>
>> Warning: no trialfun was specified, using ft_trialfun_general
>> > In ft_definetrial at 135
>> evaluating trialfunction 'ft_trialfun_general'
>> reading the events from '27CW1.RAW'
>> the following events were found in the datafile
>> event type: 'trigger' with event values: 'DIN1' 'DIN2' 'DIN4' 'DIN8'
>> no trials have been defined yet, see FT_DEFINETRIAL for further help
>> found 750 events
>> created 0 trials
>> the call to "ft_definetrial" took 4 seconds
>>
>> Then I try to use the trialfun, I get the same error
>>
>> >> cfg = [];
>> >> cfg.dataset = '27CW1.RAW';
>> >> cfg.trialfun = 'trialfun_bit2dec';
>> >> cfg.trialdef.eventtype = '?';
>> >> cfg=ft_definetrial(cfg);
>> evaluating trialfunction 'trialfun_bit2dec'
>>
>> Attempted to access trl(:,1); index out of bounds because size(trl)=[0,0].
>>
>> Error in trialfun_bit2dec (line 66)
>> idx = any(diff(trl(:,1),1,1),2);
>>
>> Error in ft_definetrial (line 169)
>> trl = feval(cfg.trialfun, cfg);
>>
>> So there is something the trialfun_bit2dec does that events cannot be
>> read anymore.
>>
>>
>> On Mon, Sep 29, 2014 at 9:01 PM, Arjen Stolk <a.stolk8 at gmail.com <mailto:
>> a.stolk8 at gmail.com>> wrote:
>>
>> Hi Ana Laura,
>>
>> Your 'trl' is empty, which it shouldn't be. So anytime you try to
>> index an element in it, it will throw an error.
>>
>> Can you check what 'event.values' you find in your dataset? And
>> compare those with any of the ones falling under the switch case
>> statement? One way to do this, is to put a debug marker (red dot)
>> at the 'end' of the first for-loop (judging from the wiki). Then
>> check the value of event(i).value, and click the continue button
>> to move on to the next. An alternative way is to put a debug
>> marker after event = ft_read_event and to instantly check all
>> values of event with "[event(:).value]".
>>
>> Goodluck,
>> Arjen
>>
>> 2014-09-30 6:20 GMT+02:00 Ana Laura Diez Martini
>> <diezmartini at gmail.com <mailto:diezmartini at gmail.com>>:
>>
>> This is my attempt of adapting the code for the new Fieldtrip
>>
>> % TRIAL DEFINITION
>>
>> cfg=[];
>> cfg.dataset = 'myfile';
>>
>> hdr = ft_read_header( 'myfile');
>> dat = ft_read_data('myfile');
>> cfg.trialfun = 'trialfun_bit2dec';
>> cfg.trialdef.eventtype = 'STATUS';
>> cfg.trialdef.eventvalue = cgrmrk; % stimulus triggers
>> cfg.trialdef.prestim = 0.2;
>> cfg.trialdef.poststim = 1;
>>
>> cfg.reref = 'yes';
>> cfg.refchannel = ['all'];
>> cfg = ft_definetrial(cfg);
>>
>> again the error is:
>>
>> Attempted to access trl(:,1); index out of bounds because
>> size(trl)=[0,0].
>>
>> Error in trialfun_bit2dec (line 66)
>> idx = any(diff(trl(:,1),1,1),2);
>>
>> Error in ft_definetrial (line 169)
>> trl = feval(cfg.trialfun, cfg);
>>
>> Error in process_ERP_1_fieldtrip (line 106)
>> cfg = ft_definetrial(cfg);
>>
>>
>>
>> On Mon, Sep 29, 2014 at 12:11 PM, Ana Laura Diez Martini
>> <diezmartini at gmail.com <mailto:diezmartini at gmail.com>> wrote:
>>
>> Dear Arjen and Nietzsche,
>>
>> I tried debugging with dbstop if error and cfg.debug =
>> 'saveonerror' and matlab points to:
>>
>> % discard the repeated values
>> idx = any(diff(trl(:,1),1,1),2);
>> trl = trl(idx,:);
>>
>> It does seem it is not reading the events. When I read
>> them with the raw data, they seem to be there, with the
>> DIN prefix. Any idea?
>>
>> On Mon, Sep 29, 2014 at 8:57 AM, Arjen Stolk
>> <a.stolk8 at gmail.com <mailto:a.stolk8 at gmail.com>> wrote:
>>
>> Hey Ana Laura,
>>
>> Seems from the error message you're getting
>>
>> "Attempted to access trl(:,1); index out of bounds
>> because size(trl)=[0,0]."
>>
>> that none of the triggers were found in your event
>> data. You might wanna check why this is happening, by
>> debugging 'trialfun_bit2dec' on your input.
>>
>> Best,
>> Arjen
>>
>>
>>
>> 2014-09-29 19:00 GMT+02:00 Ana Laura Diez Martini
>> <diezmartini at gmail.com <mailto:diezmartini at gmail.com>>:
>>
>>
>> Thank you again Nietzsche!!
>>
>> Yes, I was referring to trialfun_bit2dec. I
>> followed your advice and I changed definetrial to
>> ft_definetrial and I confirm the function was
>> added to my paths. After doing this, the error I
>> get is:
>>
>> Attempted to access trl(:,1); index out of bounds
>> because size(trl)=[0,0].
>>
>> Error in trialfun_bit2dec (line 66)
>> idx = any(diff(trl(:,1),1,1),2);
>>
>> Error in ft_definetrial (line 169)
>> trl = feval(cfg.trialfun, cfg);
>>
>> Error in process_ERP_1_fieldtrip (line 95)
>> cfg = ft_definetrial(cfg);
>>
>> This is again the trial definition part in which I
>> think I added what I think are useless lines but I
>> was just trying to make it run it.
>>
>> % TRIAL DEFINITION
>> cfg=[];
>> cfg.filename = ['myfolders/subject.RAW'];
>> cfg.headerfile = ['myfolders/subject.RAW'];
>> cfg.dataset = ['myfolders/subject.RAW'];
>> cfg.trialfun = 'trialfun_bit2dec'; %% trialfun
>> definition
>> cfg.trialdef.eventtype = 'STATUS';
>> cfg.trialdef.eventvalue = cgrmrk; % stimulus
>> triggers
>> cfg.trialdef.prestim = 0.2; % latency in seconds
>> cfg.trialdef.poststim = 1; % latency in seconds
>> cfg = ft_definetrial(cfg);
>>
>> trl = cfg.trl;
>> cfg=[];
>> cfg.dataset = ['myfolders/subject.RAW'];
>> cfg.trl = trl;
>> cfg.reref = 'yes';
>> cfg.refchannel = ['all'];
>>
>> Unfortunately using this function is crucial to my
>> analysis because I would like to use only
>> Fieldtrip to analyse all my data. Thank you for
>> taking all this time.
>>
>> On Sun, Sep 28, 2014 at 10:53 PM, Lam, Nietzsche
>> <n.lam at fcdonders.ru.nl
>> <mailto:n.lam at fcdonders.ru.nl>> wrote:
>>
>> Hi again Ana Laura,
>>
>> One other thing that I thought of was to make
>> sure that the function "trialfun_bit2dec" is
>> added to your paths in matlab, so that
>> ft_definetrial can find this function.
>>
>> By updating your fieldtrip to the most recent
>> version "trialfun_bit2dec" is *not* included.
>> So you'll need to store that as a separate .m
>> file in a location that can be accessed by the
>> paths set in matlab.
>>
>> Nietzsche
>>
>> ----- Original Message -----
>> > From: "Ana Laura Diez Martini"
>> <diezmartini at gmail.com
>> <mailto:diezmartini at gmail.com>>
>> > To: "FieldTrip discussion list"
>> <fieldtrip at science.ru.nl
>> <mailto:fieldtrip at science.ru.nl>>
>> > Sent: Saturday, 27 September, 2014 7:18:25 PM
>> > Subject: Re: [FieldTrip] Where to add a
>> trialfun?
>> > Thank you Nietzsche!
>> >
>> >
>> > I added it where you suggested and now this
>> is the error I get:
>> >
>> >
>> >
>> > Error using feval
>> > Invalid function name 'trialfun_bit2dec(cfg)'.
>> >
>> >
>> > Error in definetrial (line 105)
>> > trl = feval(cfg.trialfun, cfg);
>> >
>> >
>> > Error in process_ERP_1_fieldtrip (line 97)
>> > cfg = definetrial(cfg);
>> >
>> >
>> > Something I was worried about is that I use
>> an old version of
>> > Fieldtrip for my scripts because I wrote
>> them long ago and this
>> > trialfun uses the new format (with
>> 'ft_s',etc.). Could this affect it
>> > in any way?
>> >
>> >
>> > Thanks again!
>> >
>> >
>> > On Fri, Sep 26, 2014 at 11:05 PM, Lam,
>> Nietzsche <
>> > n.lam at fcdonders.ru.nl
>> <mailto:n.lam at fcdonders.ru.nl> > wrote:
>> >
>> >
>> > Hi Ana Laura,
>> >
>> > In general, you need to determine which
>> trial function (Trialfun) to
>> > use when using definetrial (see this tutorial:
>> >
>> http://fieldtrip.fcdonders.nl/
>> tutorial/preprocessing
>> under "do the
>> > trial definition for the fully incongruent
>> (FIC) condition).
>> >
>> > Please try adding this: "cfg.trialfun =
>> 'trialfun_bit2dec(cfg)". to
>> > your code before calling definetrial (see
>> below).
>> >
>> > % TRIAL DEFINITION
>> > cfg=[];
>> > cfg.filename = ['my/folders/', subject, '.RAW'];
>> > cfg.headerfile = ['my/folders/', subject,
>> '.RAW'];
>> >
>> > cfg.trialdef.eventtype = 'STATUS';
>> > cfg.trialdef.eventvalue = cgrmrk;
>> > cfg.trialdef.prestim = 0.2;
>> > cfg.trialdef.poststim = 1;
>> > cfg.trialdef.eventtype=?;
>> > cfg.trialfun = 'trialfun_bit2dec(cfg) %%
>> trialfun definition
>> > cfg = definetrial(cfg);
>> >
>> >
>> > As an addition note: based on your error
>> message, it seemed that the
>> > problem was in the function
>> trialfun_bit2dec. However, from the code
>> > you showed us, you haven't referenced/called
>> this function. I was
>> > wondering if the code you provide
>> corresponded to the code that
>> > created your error message? I'm guessing you
>> ran [trl]
>> > =trialfun_bit2dec(cfg) directly (i.e. not
>> via definetrial). In which
>> > case, it was looking for
>> cfg.trialdef.eventtype. You can call
>> > trialfun_bit2dec as long as you have all the
>> relevant information in
>> > the cfg (which is in the code you showed
>> us). Hope this helps.
>> >
>> > Best,
>> > Nietzsche
>> >
>> >
>> >
>> >
>> >
>> >
>> > ----- Original Message -----
>> > > From: "Ana Laura Diez Martini" <
>> diezmartini at gmail.com
>> <mailto:diezmartini at gmail.com> >
>> > > To: "FieldTrip discussion list" <
>> fieldtrip at science.ru.nl
>> <mailto:fieldtrip at science.ru.nl> >
>> > > Sent: Saturday, 27 September, 2014 2:42:21 AM
>> > > Subject: [FieldTrip] Where to add a trialfun?
>> > > Hello all! I'm having a simple problem. I
>> want to add this trialfun:
>> > >
>> > >
>> http://fieldtrip.fcdonders.nl/
>> faq/how_can_i_transform_trigger_values_from_bits_to_
>> decimal_representation_with_a_trialfun
>> > >
>> > >
>> > >
>> > > I get this error:
>> > >
>> > >
>> > >
>> > > Reference to non-existent field 'trialdef'.
>> > >
>> > >
>> > > Error in trialfun_bit2dec (line 52)
>> > > if strcmp(event(i).type,
>> cfg.trialdef.eventtype)
>> > >
>> > >
>> > > I'm quite sure it's because I'm not
>> writing it in the correct part
>> > > of
>> > > my script. This is my trial definition
>> part. Where should I add it
>> > > and
>> > > how should I write the line?
>> > >
>> > >
>> > >
>> > > % TRIAL DEFINITION
>> > > cfg=[];
>> > > cfg.filename = ['my/folders/', subject,
>> '.RAW'];
>> > > cfg.headerfile = ['my/folders/', subject,
>> '.RAW'];
>> > >
>> > >
>> > > cfg.trialdef.eventtype = 'STATUS';
>> > > cfg.trialdef.eventvalue = cgrmrk;
>> > > cfg.trialdef.prestim = 0.2;
>> > > cfg.trialdef.poststim = 1;
>> > > cfg.trialdef.eventtype=?;
>> > >
>> > >
>> > > cfg = definetrial(cfg);
>> > >
>> > >
>> > > trl = cfg.trl;
>> > > cfg=[];
>> > > cfg.dataset = ['my/folders/', subject,
>> '.RAW'];
>> > > cfg.trl = trl;
>> > > cfg.reref = 'yes';
>> > > cfg.refchannel = ['all'];
>> > >
>> > >
>> > > THANKS!
>> > >
>> _______________________________________________
>> > > fieldtrip mailing list
>> > > fieldtrip at donders.ru.nl
>> <mailto:fieldtrip at donders.ru.nl>
>> > >
>> http://mailman.science.ru.nl/
>> mailman/listinfo/fieldtrip
>> >
>> > --
>> > Nietzsche H.L. Lam, MSc
>> > PhD Candidate
>> >
>> > Max Planck Institute for Psycholinguistics
>> > Wundtlaan 1, 6525 XD Nijmegen, The Netherlands
>> >
>> > Donders Institute for Brain, Cognition and
>> Behaviour,
>> > Centre for Cognitive Neuroimaging,
>> > Kapittelweg 29, 6525EN Nijmegen, The Netherlands
>> >
>> > n.lam at fcdonders.ru.nl
>> <mailto:n.lam at fcdonders.ru.nl>
>> > +31-24-3668219 <tel:%2B31-24-3668219>
>> >
>> >
>> > neurobiologyoflanguage.com
>> <http://neurobiologyoflanguage.com>
>> > _______________________________________________
>> > fieldtrip mailing list
>> > fieldtrip at donders.ru.nl
>> <mailto:fieldtrip at donders.ru.nl>
>> >
>> http://mailman.science.ru.nl/
>> mailman/listinfo/fieldtrip
>> >
>> >
>> > _______________________________________________
>> > fieldtrip mailing list
>> > fieldtrip at donders.ru.nl
>> <mailto:fieldtrip at donders.ru.nl>
>> >
>> http://mailman.science.ru.nl/
>> mailman/listinfo/fieldtrip
>>
>> --
>> Nietzsche H.L. Lam, MSc
>> PhD Candidate
>>
>> Max Planck Institute for Psycholinguistics
>> Wundtlaan 1, 6525 XD Nijmegen, The Netherlands
>>
>> Donders Institute for Brain, Cognition and
>> Behaviour,
>> Centre for Cognitive Neuroimaging,
>> Kapittelweg 29, 6525EN Nijmegen, The Netherlands
>>
>> n.lam at fcdonders.ru.nl
>> <mailto:n.lam at fcdonders.ru.nl>
>> +31-24-3668219 <tel:%2B31-24-3668219>
>>
>>
>> neurobiologyoflanguage.com
>> <http://neurobiologyoflanguage.com>
>> _______________________________________________
>> fieldtrip mailing list
>> fieldtrip at donders.ru.nl
>> <mailto:fieldtrip at donders.ru.nl>
>> http://mailman.science.ru.nl/
>> mailman/listinfo/fieldtrip
>>
>>
>>
>> _______________________________________________
>> fieldtrip mailing list
>> fieldtrip at donders.ru.nl
>> <mailto:fieldtrip at donders.ru.nl>
>> http://mailman.science.ru.nl/
>> mailman/listinfo/fieldtrip
>>
>>
>>
>> _______________________________________________
>> fieldtrip mailing list
>> fieldtrip at donders.ru.nl <mailto:fieldtrip at donders.ru.nl>
>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>>
>>
>>
>>
>> _______________________________________________
>> fieldtrip mailing list
>> fieldtrip at donders.ru.nl <mailto:fieldtrip at donders.ru.nl>
>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>>
>>
>>
>> _______________________________________________
>> fieldtrip mailing list
>> fieldtrip at donders.ru.nl <mailto:fieldtrip at donders.ru.nl>
>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>>
>>
>>
>>
>> _______________________________________________
>> fieldtrip mailing list
>> fieldtrip at donders.ru.nl
>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>>
>
>
> --
> Jörn M. Horschig
> PhD Student
> Donders Institute for Brain, Cognition and Behaviour
> Centre for Cognitive Neuroimaging
> Radboud University Nijmegen
> Neuronal Oscillations Group
> FieldTrip Development Team
>
> P.O. Box 9101
> NL-6500 HB Nijmegen
> The Netherlands
>
> Contact:
> E-Mail: jm.horschig at donders.ru.nl
> Tel: +31-(0)24-36-68493
> Web: http://www.ru.nl/donders
>
> Visiting address:
> Trigon, room 2.30
> Kapittelweg 29
> NL-6525 EN Nijmegen
> The Netherlands
>
>
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20141005/18c7ee72/attachment-0002.html>
More information about the fieldtrip
mailing list