Closed jelmerborst closed 1 year ago
I do not see the problem you're pointing at as the code seems to do what is expected. Can you specify more what the actual vs. the expected behavior of those variables is?
Nm, turned out there's an issue with using mne.filter.resample on matrices containing NaNs. After solving that, this bug turned out to be non-existent.
At least when fitting multiple subjects,
mean_d
anddurations
of the hmp model are based on the max duration (ie number of samples in the matrix) of each subject. They should be based on the mean of each subject.This might be caused by that
data.unstack()
on line 60 ofmodels.py
gives the same number of epochs as there are subjects.