Currently, the terminator node incorrectly terminating a lot of the change_u radiation runs. Because it spikes due to unexpected excitation. This can be resolved by:
Options
[ ] Computing the maximum runtime using Neumann.
[ ] Computing the minimum runtime using Neumann.
[ ] Let it run for at least the minimum runtime, and start checking if the network is completed by:
[ ] Seeing whether a neuron spiked. (and returning a boolean from the simulator)
[ ] Seeing whether a neuron spiked in the spike data of the last timestep
[ ] Having a Terminator node that gets excited until the network is done.
[ ] Having a Terminator node that gets inhibited until the network is done.
[ ] Only terminating the network if all Terminator neurons are done (This is not possible for both excitatory and inhibitory).
Complexity: O something.
Options
Computing the minimum and maximum runtime in neuron countdown and using that to determine whether the network is done.
Currently, the terminator node incorrectly terminating a lot of the
change_u
radiation runs. Because it spikes due to unexpected excitation. This can be resolved by:Options
Options
Computing the minimum and maximum runtime in neuron countdown and using that to determine whether the network is done.