model.modules.sli.SLI

model.modules.sli.SLI#

class SLI(p: SLIParameters = (tensor(200.), tensor(100.), tensor(0.), tensor(1.)), **kwargs)[source]#

Bases: SNN

A neuron layer that wraps a saturable-leaky-integrator SLICell in time.

The layer iterates over the _outer_ dimension of the input.

Parameters:

  • p (float, optional) – parameters of the leaky integrator

  • dt – integration timestep to use

Methods

initial_state

Attributes

training

initial_state(input_tensor: Tensor) → SLIState[source]#
Contents