superneuromat.SNN.reset

SNN.reset()

Reset the SNN’s neuron states, refractory periods, spike train, and input spikes.

Equivalent to:

snn.reset_neuron_states()
snn.zero_refractory_periods()
snn.clear_spike_train()
snn.clear_input_spikes()
snn.restore()

Warning

This method does not reset the synaptic weights or STDP parameters. SuperNeuroMAT also does not automatically store the initial neuron state values, such as the initial_state and refractory_state parameters of create_neuron(). When reset() is called, if those states are not memoized, each neuron’s charge state will be set to its reset state in neuron_reset_states, and the refractory countdown in neuron_refractory_periods_state will be set to zero.

If this is not desirable, consider manually copying the parameters you care about so you can assign them later, or using memoize() to store a snapshot to return to when restore() or reset() is called, or manually calling only the individual functions that you need (shown above).

See Resetting the SNN for more information.

See also

reset_neuron_states

Reset the charge states of all neurons to their reset values.

zero_refractory_periods

Reset the refractory period countdowns to zero.

clear_spike_train

Delete all recorded spike trains.

clear_input_spikes

Delete all queued input spikes.

restore

Restore the model variables to their memoized states.