jaxsnn.event.trajectory

Classes

Any(*args, **kwargs)	Special type indicating an unconstrained type.
EventPropSpike(time, idx, current)
InputQueue(spikes, head)
LIFState(V, I)
StepState(neuron_state, spike_times, …)

Functions

jaxsnn.event.trajectory.filter_spikes(input_spikes: jaxsnn.event.types.EventPropSpike, prev_layer_start: int) → jaxsnn.event.types.EventPropSpike

Filters the input spikes by ensuring only the spikes from the previous layer are kept.

jaxsnn.event.trajectory.trajectory(step_fn: Callable[[Tuple[jaxsnn.event.types.StepState, Union[jaxsnn.event.types.WeightInput, jaxsnn.event.types.WeightRecurrent], int], int], Tuple[Tuple[jaxsnn.event.types.StepState, Union[jaxsnn.event.types.WeightInput, jaxsnn.event.types.WeightRecurrent], int], jaxsnn.event.types.EventPropSpike]], size: int, n_spikes: int) → Callable[[int, Union[jaxsnn.event.types.WeightInput, jaxsnn.event.types.WeightRecurrent], jaxsnn.event.types.EventPropSpike], jaxsnn.event.types.EventPropSpike]

Evaluate the step_fn until n_spikes have been simulated.

Uses a scan over the step_fn to return an apply function