spark.nn.components.synapses.base

Attributes

ConfigT

Classes

SynanpsesOutput	Generic synapses model output spec.
SynanpsesConfig	Abstract synapse model configuration class.
Synanpses	Abstract synapse model.

Module Contents

class spark.nn.components.synapses.base.SynanpsesOutput

Bases: TypedDict

Generic synapses model output spec.

Initialize self. See help(type(self)) for accurate signature.

currents: spark.core.payloads.CurrentArray

class spark.nn.components.synapses.base.SynanpsesConfig(__skip_validation__=False, **kwargs)

Bases: spark.nn.components.base.ComponentConfig

Abstract synapse model configuration class.

Parameters:

__skip_validation__ (bool)

spark.nn.components.synapses.base.ConfigT

class spark.nn.components.synapses.base.Synanpses(config=None, **kwargs)

Bases: spark.nn.components.base.Component, Generic[ConfigT]

Abstract synapse model.

Note that we require the kernel entries to be in pA for numerical stability, since most of the time we want to run in half-precision. However somas expect the current in nA so we need to rescale the output.

Init:

Input:

spikes: SpikeArray

Output:

currents: CurrentArray

Parameters:

config (ConfigT | None)

abstractmethod get_kernel()

Return type:

spark.core.payloads.FloatArray

abstractmethod set_kernel(new_kernel)

Parameters:

new_kernel (spark.core.payloads.FloatArray)

Return type:

None

__call__(spikes)

Compute synanpse’s currents.

Parameters:

spikes (spark.core.payloads.SpikeArray)

Return type:

SynanpsesOutput