Page 2 of 2

battery

Posted: 01 Nov 2010, 07:34
by gschalk
Hi,

The battery status is not written to the file. In general, no BrainProducts-specific information should influence the operation of the P300Classifier. The only thing I can think of is that the low battery status changes the way the data are acquired (perhaps one or more channels contain only zeros, etc.), and that the P300Classifier is not robust to this sort of data. Thus, it would be important for us to get these "bad" data so we can investigate the problem.

Thanks,
Gerv

Posted: 07 Jan 2011, 10:02
by faisal-awan
hi

i need to do ...Offline Analysis of P300 signal ..... I donot know that src/contrib./Tools/P300GUI/P300_GUI.m , where i get this GUI . i my version of BCI2000 ... i donot have this .


Please help me in this.




Regards

P300GUI

Posted: 07 Jan 2011, 10:05
by gschalk
The P300 GUI is part of BCI2000 V2. V3 comes with the P300Classifier instead, which performs the same functions.

Both versions are available on SVN as described on doc.bci2000.org.

Gerv

Re: battery

Posted: 01 Mar 2011, 10:17
by tcetin
gschalk wrote:Hi,

The battery status is not written to the file. In general, no BrainProducts-specific information should influence the operation of the P300Classifier. The only thing I can think of is that the low battery status changes the way the data are acquired (perhaps one or more channels contain only zeros, etc.), and that the P300Classifier is not robust to this sort of data. Thus, it would be important for us to get these "bad" data so we can investigate the problem.

Thanks,
Gerv
Hey,
I have done some more experiments and still come up to the same problem - When we record a training session right after the Amp is switched on with the fully chrged battery we can use it in the P300Classifier, but all session afterwards we cannot use. When loading them in the Classifier it just crashes. I connected our Amp to the mains and tried that, but then we get the message "Data must be arranged in Rows and Columns".

I will send you one of the Bad files (i.e. a file that caused the P300Classifier to crash) and a Good file (one that is recorded just after turning the amp on an we could use for classification) via email.

Thanks,
Timur

--
Dipl.-Biol. Timur Cetin
University of Bremen
Dept. of Psychology and Cognitive Research