Alexandre Gramfort
Alexandre Gramfort
you don't correct for ECG/EOG with ICA? just EOG and big artifacts with reject? how did you set the reject thresholds?
ok so I understand that for your use case autoreject was removing too much data.
I am afraid this will change a bit the results of our paper. I don't have time to rerun all to check if it's the case... :(
ICA is fitted jointly so I guess it will have some impact. Hopefully tiny...
I think ICA code makes 2 copies of the data
no objection
have a look at https://pactools.github.io > Message ID: ***@***.***> >
ping @TomDLT who might be interested in giving a hand here
maybe we can just extend plot_sensors_connectivity to have a 2D matplotlib version? >
Sounds good 👌