lcnhappe/happe

Harvard Automated Pre-Processing Pipeline for EEG

happe's Issues

.mat file imported exceeds maximum array size preference

I am running the latest version of happe. But when this HAPPE is running for the sample datasets mentioned in the paper. But I ran into this error when ran for the 'baselineEEG001.mat' file. Below is the error.

Error using HAPPE_pipeline_v1_0 (line 259)
Requested 129x87250x1000 (41.9GB) array exceeds maximum array size preference. Creation of arrays greater than this
limit may take a long time and cause MATLAB to become unresponsive.

Video of error attached.

Uploading happer_error.zip…

Issue related to ICA decomposition

Dear HAPPE developers,

I am encountering an issue with one of the subjects in running the ICA decomposition.

The file contains 124 channels.
HAPPE removes 2 bad channels.
Performing ICA gives me the following result:

Warning: fixing rank computation inconsistency (111 vs 110) most likely because running under Linux 64-bit Matlab
Attempting to convert data matrix to double precision for more accurate ICA results.
Data rank (111) is smaller than the number of channels (122).

Does this suggest that perhaps there are residual bad channels in the data? Due to which the rank is much smaller than the total number of channels?

Kindly suggest a possible workaround for this problem.
Thank you very much,

Velu

Link to Happe 3.0 / Mark non-maintained

Hi!
I had to actually search for the newest HAPPE version, the link in the frontiers paper points here.

Maybe a note in the README would be helpful, where users can find the newest version :)

Cheerio

PS: For those looking for the newest version, you can find it here: https://github.com/PINE-Lab/HAPPE

eml_invariant function? which toolbox?

Hey,

I encountered a problem while using your pipeline. Namely, there is a function that I'm not able to find out which toolbox it belongs. I have all required toolboxes that you specified in your README file.

The function I'm talking about is eml_invariant.

Here is error message from MATLAB:

"
Undefined function or variable 'eml_invariant'.

Error in coder.internal.parseParameterInputs (line 82)
eml_invariant(nargin >= 2, ...

Error in dwt (line 33)
pstruct = coder.internal.parseParameterInputs(parms,poptions, ...

Error in wavedec (line 42)
[xv,d] = dwt(xv,Lo_D,Hi_D); % decomposition

Error in ddencmp (line 77)
[c,l] = wavedec(x,1,'db1');

Error in wICA (line 94)
[thresh,sorh,~] = ddencmp('den','wv',sig); % get automatic threshold value

Error in happe_acz (line 350)
[wIC, A, W, IC] = wICA(EEG,'runica', 1, 0, [], 5);
"

I use Windows 10 Pro and Matlab 2018a.

Thanks in advance for your answer.

Best,
Artur

empty variables and missing assignments in wICA

Hey,

great idea to put together these tools and prepare a fully automated pipeline.

I tried to use it and encounter problems from the beginning.

wICA function doesn't return any values
I think the problem is with assignments of variable: IC
it's simply empty.

the function 'runamica' doesn't calculate OUTEEG.icaact
As far as I know you are missing this in wICA:

OUTEEG.icaact = (OUTEEG.icaweights*OUTEEG.icasphere)*OUTEEG.data(OUTEEG.icachansind,:);
Let me know if im not correct

Did you test your pipeline?
I'm looking forward to your answer.

Best,
Artur

Channel Location & Event-Related information gets deleted after HAPPE Processing

Dear HAPPE developers,

After pre-processing my data using HAPPE, I could not find the channel labels and event-related information anymore.
Could you please point where should the fix be done?

Thank you,

Velu