[Eeglablist] Applying ASR / ICA to NeurofeedbackLab
Arnaud Delorme
adelorme at ucsd.edu
Wed Nov 13 13:00:13 PST 2024
Is it possible that the file "data_nfblab_2024-11-12_11-14_asr_filter.mat" does not contain ASR if the ASR option was turned off and then you ran the program again with the option turned on, selected this baseline file, and it could not find the pre-saved information? That would explain the problem. It could be something else as well. If that’s the case, please submit an issue on GitHub.
https://urldefense.com/v3/__https://github.com/arnodelorme/neurofeedbacklab/issues__;!!Mih3wA!DiMKMhS8vhvnYeAHn0MCMBBt4nCzju8UXdGzbv1XdJEsLmTyTqti-tZzHDfJaEC0A_1q5tmcREv9EcQwYhD1Qt7I$
Make sure you use the latest version on Github.
Best wishes,
Arno
> On Nov 13, 2024, at 08:06, Sam Roberts via eeglablist <eeglablist at sccn.ucsd.edu> wrote:
>
> Hi,
> I'm trying to get a protocol sorted using NeurofeedbackLab<https://urldefense.com/v3/__https://github.com/arnodelorme/neurofeedbacklab__;!!Mih3wA!ExiWeSW5qUOBSoonPj8y-PyHdBne9exic1cTXZEOJYjGGl1RTFSICB_BOFDjqiwZgSYRzkJAqx9Vbq2Oa8k-OX5hK_F6v_6IRceV-KEov37C$ >, I haven't done much beyond downloading and successfully replicating the YouTube demos, but as I move into my own system I am having struggles in getting the real-time ICA/ASR to work
>
> The only change I've made is setting asrFlag and icaFlag to true in the nfblab_setfields script and these errors come back:
>
> %ASR
> Starting new session...
> Loading baseline ASR file data_nfblab_2024-11-12_11-14_asr_filter.mat...
> eeg_checkset note: upper time limit (xmax) adjusted so (xmax-xmin)*srate+1 = number of frames
> Feedback "Streaming" sent to client (if any)
> Dot indexing is not supported for variables of this type.
>
> Error in asr_process (line 108)
> [T,M,A,B] = deal(state.T,state.M,state.A,state.B);
> ^^^^^^^
> Error in nfblab_process (line 576)
> [chunkFilt(nonEventChans,:), stateAsr]= asr_process(chunkFilt(nonEventChans,:), g.input.srateHardware, stateAsr, [],[],[],[],size(g.input.chans));
>
>
>
> %ICA
> Starting new session...
> Loading baseline ASR file data_nfblab_2024-11-12_11-14_asr_filter.mat...
> eeg_checkset note: upper time limit (xmax) adjusted so (xmax-xmin)*srate+1 = number of frames
> Feedback "Streaming" sent to client (if any)
> Error using *
> Incorrect dimensions for matrix multiplication. Check that the number of columns in the first matrix matches the number of rows in
> the second matrix. To operate on each element of the matrix individually, use TIMES (.*) for elementwise multiplication.
>
> Error in nfblab_process (line 581)
> icaAct = g.preproc.icaWeights(g.preproc.icaRmInd,:)*chunkFilt(nonEventChans,:);
>
> I understand what 'dot indexing not supported' and 'incorrect dimensions for matrix multiplication' means in a general MATLAB sense, but I don't understand what else needs to be done to make this work.
>
> I also tried to import the baseline .mat file to EEGLab as maybe you'd need to calculate the ICA weights offline, but EEGLab couldn't import the data from the .mat file
> [cid:838c746c-c9e0-4c39-b6e9-c60602cd02a9]
>
> Does anyone have experience with NeurofeedbackLab and can help me in getting the real-time ASR and ICA to work?
>
>
> Best Regards,
>
>
>
> -Sam Roberts
>
> PhD Student
>
> Division of Psychology, Communication & Human Neuroscience
>
> The University of Manchester
>
> Connect with me on LinkedIn<https://urldefense.com/v3/__https://www.linkedin.com/in/sam-roberts-5b3261209/__;!!Mih3wA!ExiWeSW5qUOBSoonPj8y-PyHdBne9exic1cTXZEOJYjGGl1RTFSICB_BOFDjqiwZgSYRzkJAqx9Vbq2Oa8k-OX5hK_F6v_6IRceV-ErcRFw7$ >
>
>
> _______________________________________________
> 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
More information about the eeglablist
mailing list