[FieldTrip] A question

[Boris Reuderink]

Further, it seems that the data is also scaled automatically, see
http://code.google.com/p/fieldtrip/source/browse/trunk/ft_componentanalysis.m?spec=svn5313&r=5313#271
.

Best,

Boris

On Thu, Feb 16, 2012 at 12:39 PM, Stephen Whitmarsh
<stephen.whitmarsh at gmail.com> wrote:
> Dear All,
>
> All true, except that ICA implemented by FT does demean the data
> before unmixing by default. Use cfg.demean = 'no' if you don't want
> to.
>
> cheers,
> stephen
>
> On 16 February 2012 12:35, Stephen Politzer-Ahles
> <politzerahless at gmail.com> wrote:
>> Hi Shayda,
>>
>> FieldTrip does not apply a filter by default, and I am not aware of any
>> other 'hidden' transformations that happen. You can specify a filter when
>> importing the data with ft_preprocessing(), and there are several different
>> kinds of filters available, which will affect your data in different ways--
>> see http://fieldtrip.fcdonders.nl/reference/ft_preprocessing. The best thing
>> to do is to import your data without using a filter and, if possible, use
>> whatever tool comes with your MEG software to import that raw data into
>> MATLAB as well (for example, for CTF this would be ds2txt), then you can
>> plot them together to make sure there are no differences--when I have tried
>> this with EEG data, fieldtrip hasn't changed them. Of course, if you want to
>> use a filter, you can also try several filters (you can either filter within
>> your system's software and then import straight to Fieldtrip, or use one of
>> Fieldtrip's filters during import) and plot them to see how they affect your
>> data.
>>
>> Best,
>> Steve Politzer-Ahles
>>
>>
>>>
>>>
>>> Message: 2
>>> Date: Wed, 15 Feb 2012 14:19:50 -0600
>>> From: Shayda Malekpour <shayda.malekpour at gmail.com>
>>> To: fieldtrip at donders.ru.nl
>>> Subject: [FieldTrip] A question
>>> Message-ID:
>>>
>>>  <CAMUfAt=cGAnjLDby96CnXi4ubV=v2RCDnM9HoH2ObG5ZT5Bdeg at mail.gmail.com>
>>> Content-Type: text/plain; charset="iso-8859-1"
>>>
>>>
>>> Dear All,
>>>
>>> I am using fieldtrip on MEG data set.
>>> I want to know if I do not remove any ICA component,  will the data be
>>>  the
>>> same as the original data, or is there is a filter somewhere in the
>>> software that changes the data? If so how much does it change it (What is
>>> their relative difference?)
>>>
>>> Thanks,
>>>
>>> Sheida
>>> -------------- next part --------------
>>> An HTML attachment was scrubbed...
>>> URL:
>>> <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20120215/ba4458d8/attachment-0001.html>
>>>
>>> ------------------------------
>>>
>>> Message: 3
>>> Date: Thu, 16 Feb 2012 09:28:43 +0100
>>> From: Eelke Spaak <eelke.spaak at donders.ru.nl>
>>> To: Email discussion list for the FieldTrip project
>>>        <fieldtrip at donders.ru.nl>
>>> Subject: Re: [FieldTrip] A question
>>> Message-ID:
>>>
>>>  <CABPNLUp8hQFKG7d1cuUQt4rkb3mfqxtqDWk4AyKrf5ZX8b3BYA at mail.gmail.com>
>>> Content-Type: text/plain; charset=ISO-8859-1
>>>
>>>
>>> Dear Sheida,
>>>
>>> You probably need to provide a bit more information before anyone is
>>> able to help you. In general, if you perform ICA, you get out
>>> component timecourses and mixing/unmixing matrices. If you keep all
>>> your components, and multiply them with the mixing matrix, you should
>>> get all your original channel timecourses out again. So: if you don't
>>> change the data, the data will not be changed :)
>>>
>>> Best,
>>> Eelke
>>>
>>> On 15 February 2012 21:19, Shayda Malekpour <shayda.malekpour at gmail.com>
>>> wrote:
>>> > Dear All,
>>> >
>>> > I am using fieldtrip on MEG data set.
>>> > I want to know if I do not remove any ICA component,? will the data be?
>>> > the
>>>
>>> > same as the original data, or is there is a filter somewhere in the
>>> > software
>>> > that changes the data? If so how much does it change it (What is their
>>> > relative difference?)
>>> >
>>> > Thanks,
>>> >
>>> > Sheida
>>> > _______________________________________________
>>> > fieldtrip mailing list
>>> > fieldtrip at donders.ru.nl
>>> > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>>>
>>>
>>>
>>> ------------------------------
>>>
>>> Message: 4
>>> Date: Thu, 16 Feb 2012 10:36:05 +0100
>>> From: "J?rn M. Horschig" <jm.horschig at donders.ru.nl>
>>> To: Email discussion list for the FieldTrip project
>>>        <fieldtrip at donders.ru.nl>
>>> Subject: Re: [FieldTrip] A problem in real-time processing of MEG
>>>        signal using field trip buffer
>>> Message-ID: <4F3CCE05.4050109 at donders.ru.nl>
>>> Content-Type: text/plain; charset="iso-8859-1"; Format="flowed"
>>>
>>> Dear Soheila,
>>>
>>> we were facing the same problem here. Presumably, there is some problem
>>> in the shared memory module that the buffer is using. Right now,
>>> however, we could not find out where the error precisely comes from, so
>>> our best work around is to restart the computer prior recording and
>>> optimize settings, such as channels and sampling/decimation rate of
>>> acq2ft. In our lab, we managed to find near-optimal settings, so that
>>> the buffer crashes only rarely (actually, I never saw it crashing when
>>> the acquisition computer was freshly restarted). Note that a lower
>>> number of channels or higher decimation does not mean that it runs more
>>> stable.  It seems to be literally a trial-and-error approach :/
>>>
>>> In order to track this down further, we need some more information, e.g.
>>> what MEG system are you using? What's your sampling rate, how many
>>> channels do you put into the buffer, and what's your decimation rate? It
>>> would be great if you could send me that information so that we can
>>> maybe try to track this down.
>>>
>>> Best regards,
>>> J?rn
>>>
>>> On 2/14/2012 5:42 PM, Soheila Samiee wrote:
>>> > Dear group,
>>> >
>>> > I use FieldTrip for real-time processing of MEG signals using shared
>>> > memory and field trip buffer in a machine other than acquisition
>>> > system. I use matlab for reading the buffer and processing the data.
>>> > During my data acquisition and sending it to other computer with
>>> > /acq2ft/ in linux, I face a problem. After sending some blocks of
>>> > data, this message: "/Internal Converter thread does not keep up with
>>> > load/" is shown in the terminal of /acq2ft/ and in my matlab session I
>>> > see an error : "ERROR: the buffer returned an error (517)". It does
>>> > not depend on sampling rate of data in acquisition system. The problem
>>> > often solve when I restart the computer which receives the data, but
>>> > after some minutes of transferring the data I see the message again.
>>> >
>>> > Would you please help me in solving this problem?
>>> > Best Wishes,
>>> > Soheila
>>>
>>> >
>>> >
>>> > _______________________________________________
>>> > fieldtrip mailing list
>>> > fieldtrip at donders.ru.nl
>>> > http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>>>
>>>
>>> --
>>> J?rn M. Horschig
>>> PhD Student
>>> Donders Institute for Brain, Cognition and Behaviour
>>> Centre for Cognitive Neuroimaging
>>> Radboud University Nijmegen
>>> Neuronal Oscillations Group
>>>
>>> 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
>>>
>>> -------------- next part --------------
>>> An HTML attachment was scrubbed...
>>> URL:
>>> <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20120216/c01be767/attachment.html>
>>>
>>> ------------------------------
>>>
>>>
>>> _______________________________________________
>>> fieldtrip mailing list
>>> fieldtrip at donders.ru.nl
>>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>>>
>>> End of fieldtrip Digest, Vol 15, Issue 17
>>> *****************************************
>>
>>
>>
>>
>> --
>> Stephen Politzer-Ahles
>> University of Kansas
>> Linguistics Department
>> http://www.linguistics.ku.edu/
>>
>> _______________________________________________
>> fieldtrip mailing list
>> 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



-- 
twitter.com/#!/breuderink | github.com/breuderink | borisreuderink.nl