Demo Notebook "simulate_surface_seeg_eeg_meg.ipynb"

[MICHELANGELO FABBRIZZI]

Dear All,

I am trying to use the Notebook "simulate_surface_seeg_eeg_meg.ipynb", I have installed last version of TVB (2.7.2) and downloaded the last version of the notebook from GitHub https://github.com/the-virtual-brain/tvb-root/blob/master/tvb_documentation/demos/simulate_surface_seeg_eeg_meg.ipynb , at the same time when I run it as it is (no modification at all) I get the below Error: could anybody help?

Appreciated, regards,

Mic

--------------------------------------------------------------------------- ValueError Traceback (most recent call last) Cell In[2], line 48 37 default_cortex.coupling_strength = local_coupling_strength 39 sim = simulator.Simulator( 40 model=oscillator, 41 connectivity=white_matter, (...) 46 simulation_length=10.0 47 ) ---> 48 sim.configure() 50 eeg, meg, seeg, _ = sim.run() File /Applications/tvb-2.7.2.app/Contents/Resources/lib/python3.10/tvb/simulator/simulator.py:272, in Simulator.configure(self, full_configure) 270 self._configure_history() 271 # Configure Monitors to work with selected Model, etc... --> 272 self._configure_monitors() 273 # Estimate of memory usage. 274 self._census_memory_requirement() File /Applications/tvb-2.7.2.app/Contents/Resources/lib/python3.10/tvb/simulator/simulator.py:433, in Simulator._configure_monitors(self) 431 # Configure monitors 432 for monitor in self.monitors: --> 433 monitor.config_for_sim(self) File /Applications/tvb-2.7.2.app/Contents/Resources/lib/python3.10/tvb/simulator/monitors.py:688, in EEG.config_for_sim(self, simulator) 687 def config_for_sim(self, simulator): --> 688 super(EEG, self).config_for_sim(simulator) 689 self._ref_vec = numpy.zeros((self.sensors.number_of_sensors,)) 690 if self.reference: File /Applications/tvb-2.7.2.app/Contents/Resources/lib/python3.10/tvb/simulator/monitors.py:576, in Projection.config_for_sim(self, simulator) 574 else: 575 full_gain = numpy.zeros((self.gain.shape[0], self.gain.shape[1] + sub_gain.shape[1])) --> 576 full_gain[:, self.rmap if using_cortical_surface else cortical_indices] = self.gain 577 full_gain[:, non_cortical_indices] = self.analytic(*src) 578 self.gain = full_gain ValueError: shape mismatch: value array of shape (62,16384) could not be broadcast to indexing result of shape (62,16500)

[MICHELANGELO FABBRIZZI]

Apologize for the bad formatting, I attach the Notebook (available at the github link above) and a screenshot of the error.

[Lia Domide]

Hi Mic,

Thank you for reporting this here.

We have a bug reported on this topic: https://req.thevirtualbrain.org/browse/TVB-3018

Unfortunately I can not give a time estimation for when it will be fixed. But we are open to contributions, in case someone finds the resources and the inspiration.

Best,

Lia.

 

[Lia Domide]

In case it helps anyone:

In a recent change here, we adjusted the  simulate_surface_seeg_eeg_meg.ipynb to use a connectivity without subcortical regions for the surface simulation, to avoid our bug TVB-3018 (which is still there for now).