[Eeglablist] 'model' problems while using BCILAB online.

[Christian Kothe]

In the online processing system, data streams and predictors are referred to by their name, and this (somewhat obscure) error comes up when one passes the stream struct instead of its name to onl_newpredictor. A patch with improved error message (and aside, auto-identification of the streams that are required for most cases) is now found in the patch sub-directory under:

ftp://sccn.ucsd.edu/pub/bcilab

Christian


On Apr 12, 2011, at 9:20 PM, marsel mano <marsel.mano at gmail.com> wrote:

> Hi everyone,
> 
> I recently started to use BCILAB 0.9 beta2b and everything is working fine when I process the data offline.
> But when it comes to online processing i can't create an online predictor with the model and the stream that I have. 
> The error is as follows:
> 
> >>onl_newpredictor('mypredictor',lastmodel,mystream);
> Undefined function or method 'evalin' for input arguments of type 'struct'.
> occured in
>   resolve_rawdata: 171
>   resolve_rawdata: 236
>   resolve_rawdata: 236
>   resolve_rawdata: 236
>   resolve_rawdata: 236
>   onl_newpredictor: 124
> ??? Error using ==> onl_newpredictor at 140
> The given model has an unexpected structure.
> 
> When I load EEGLAB datasets I get the following message:
> 
>   The loaded EEGLAB set is lacking an online expression; assuming it contains unfiltered data.
>    If it contains filtered data, however, BCI models derived from it will likely not be online-capable.  
> 
> then I use bci_train to get the model which by the way has some online expressions on tracking field. Anyway when it comes to creating a predictor it shows the above error.
> 
> I tried other training dataset types (".dat", ".bdf",".vhdr") but still nothing.
> I am not sure if it is the problem with the training dataset or training approach, and I have tried training different approaches as well but still no results.  
> 
> Can anyone tell me how to solve this issue?
> If anyone could share some suitable EEGLAB sets successfully used for training online capable models, would be appreciated.
> 
> With Best Regards
> 
> -- 
> Marsel Mano
> PhD Candidate (University Of Toyama)
> Toyama, Japan
> +81 80 4253 8833
> 
> _______________________________________________
> 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/20110414/744265e2/attachment.html>