Hi,
I managed to read the TRF output from Matlab into MNE and eventually apply the meta-data (info) from a raw file. I now have an evoked object with 0.9s length and 129 channels. evoked.plot() shows me pretty much the expected curves. I can plot the sensor localization and the whole thing looks like a brain.
evoked.pick_types(meg=“grad”).plot_topo(color=“r”, legend=True)
this also shows reasonable results.
How do I perform source localization on these evoked data?
I assume the problem is the computation of the noise covariance.
What procedure would be required in this case?
thanks,
Dennis