Updating Custom Stimulus for New Version of TVB

[canthony....@gmail.com]

Hi all,

I defined a custom closed-loop stimulus class in a previous version of TVB (1.5.3, I think). However, since getting a new computer, I have switched to a newer version of TVB (1.5.8), and it seems that the custom stimulus class no longer works. I included my code and the error that I received below.

If the average activity of a particular state variable at each node exceeds a value, the code defines a waveform and iteratively inputs a point of that waveform to the stimulator at each following time point. After a certain period of stimuli, a pause period is defined that prevents stimulation for a certain number of time points.

Does that make sense, and are there any recommendations on how to fix this?

Thank you,

Charles

CODE:

# Create Stimulation Class

class Stim(patterns.StimuliRegion):

    def __init__(self, *args, **kwargs):

        super(Stim, self).__init__(*args, **kwargs)

        self.threshold1 = -.8 # threshold for mean activity of state variable 0

        self.threshold2 = -.5 # threshold for mean activity of state variable 1

        self.T1 = 100

        self.T = self.T1 #int(self.T1*(1/int_time_step))

        self.stim_initiated = 0;  #has stimulation been initiated?; controls whether import first stim value/start stim or output later stim values

        self.timeleft2stim = self.T-1; #length of stim period minus one; a decreasing indicator of whether stimulation should occur

        self.stim_idx = 0;  #index of stimulus to output

        self.pause = 0; #this is used to create a pause to let the amplitude of the model return to interictal after stimulation

    def __call__(self, time, space=None):#nodes):

        n_node = self.state.shape[1]

        self.stimulus = np.zeros(n_node)

        if self.pause == 0: # if it is possible to stimulate at this time (i.e. the pause indicator is not active)

            if self.stim_initiated == 0: # if stimulation is not ongoing

                if time > 2e3 and mean(self.state[0, epileptogenic_area_indices, 0]) > self.threshold1 and mean(self.state[1, epileptogenic_area_indices, 0]) > self.threshold2:

                    self.freq = 10 # number of stimuli per 100 time points (T)

                    self.amp = 9 # amplitude of stimulation

                    self.t = np.linspace(0,1,self.T,endpoint = False) # define time indices

                    self.sig = ((sc.signal.square(2*np.pi*self.freq*self.t)+1)/2)*self.amp # define waveform that you can iterate over

                    self.stimulus[epileptogenic_area_indices] = self.sig[self.stim_idx] # feed value from waveform into stimulus variable

                    self.stim_initiated = 1 # signal that stimulation has been turned on

                    self.timeleft2stim -= 1 # decrease the remaining length of stimulation time by 1 point

            elif self.stim_initiated == 1: # if stimulation is ongoing

                self.stim_idx +=1 # advance the index of the stimulation waveform that will be passed to the stimulator

                self.timeleft2stim -=1

                if self.timeleft2stim > 0: # if the amount of time left to stim is greater than 0

                    self.stimulus[epileptogenic_area_indices] = self.sig[self.stim_idx] # pass the next index of the signal to the stimulus

                else:

                    self.stim_initiated = 0

                    self.timeleft2stim = self.T-1 #reset stim length value

                    self.stim_idx = 0 # reset stimulus index

                    self.pause = 400 # set pause to 100 time points (i.e. no stimulus will occur for 400 time points)

                    self.stimulus[epileptogenic_area_indices] = 0 # turn the stimulus value to 0

        else: # if stimulation is on pause

            self.pause -= 1 # reduce the remaining pause duration

            self.stimulus[epileptogenic_area_indices] = 0 # make sure the stimulus value is set to 0

        return self.stimulus # maybe try self.eqn_t

    def set_state(self, state):

        self.state = state

    def configure_time(self, t):

        pass

stim = Stim()

stim.configure()

ERROR:

---------------------------------------------------------------------------

TraitAttributeError                       Traceback (most recent call last)

<ipython-input-9-5a00b4aa5ed7> in <module>

      2 

      3 stim = Stim()

----> 4 stim.configure()

      5 

      6 hiss = noise.Additive(nsig = numpy.array([0.003, 0.003, 0., 0.003, 0.003, 0.]))

~\anaconda3\envs\TVB_env\lib\site-packages\tvb\basic\neotraits\_core.py in configure(self, *args, **kwargs)

    222         Override to compute uninitialized state of the class.

    223         """

--> 224         self.validate()

    225 

    226 

~\anaconda3\envs\TVB_env\lib\site-packages\tvb\basic\neotraits\_core.py in validate(self)

    214             # read all declarative attributes. This will trigger errors if they are

    215             # in an invalid state, like beeing required but not set

--> 216             getattr(self, k)

    217 

    218 

~\anaconda3\envs\TVB_env\lib\site-packages\tvb\basic\neotraits\_attr.py in __get__(self, instance, owner)

    168         if self.required and value is None:

    169             raise TraitAttributeError('required attribute referenced before assignment. '

--> 170                                       'Use a default or assign a value before reading it', attr=self)

    171         return value

    172 

TraitAttributeError: required attribute referenced before assignment. Use a default or assign a value before reading it

  attribute tvb.datatypes.patterns.StimuliRegion.connectivity = Attr(field_type=<class 'tvb.datatypes.connectivity.Connectivity'>, default=None, required=True)

[Lia Domide]

Hi,

The error you get says that a Connectivity is required on stim = Stim() object, before it can be configured.

Could you add something like stim.connectivity = Connectivity.from_file() # or another Connectivity instance, as you would be using in your simulation

Best,

Lia.

--
You received this message because you are subscribed to the Google Groups "TVB Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tvb-users+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/tvb-users/223f5e28-d561-4439-b618-bc084fc8df09n%40googlegroups.com.