jaxsnn.event.modules.hx.linear

Classes

DLSGlobal

Experiment(topology, *args, …)

Hardware experiment class for executing experiment on BrainScaleS-2

Parameter

alias of jax.Array

Population(layer_idx, n_events, n_hw_spikes, …)

Base class for all populations

Projection(generator, int], …)

Event-driven projection module

Synapse(layer_idx, source_population, …)

Synapse layer

Functions

jaxsnn.event.modules.hx.linear.HXLinear(mean: float = 0.5, std: float = 2.0, min_delay: float = 0.0, weight_scale: float = 1.0, chip_coordinate: Optional[pyhalco_hicann_dls_vx_v3.DLSGlobal, None] = None, transform: Callable = <function linear_saturating>) → jaxsnn.event.types.Projection

Creates a synapse initialization function and associated parameters.

Parameters

  • mean – Mean value for initializing synaptic weights.

  • std – Standard deviation for initializing synaptic weights.

  • min_delay – Minimum allowable synaptic delay.

  • weight_scale – Scaling factor for synaptic weights.

  • chip_coordinate – Chip coordinate for hardware execution.

  • transform – Transformation function applied to synaptic weights.

Returns

Projection instance containing generator function and parameters.

jaxsnn.event.modules.hx.linear.linear_saturating(weight: jax.Array, scale: float, min_weight: float = - 63.0, max_weight: float = 63.0, as_int: bool = True) → jax.Array

Scale all weights according to:

w <- clip(scale * w, min_weight, max_weight)

Parameters

  • weight – The weight array to be transformed.

  • scale – A constant the weight array is scaled with.

  • min_weight – The minimum value, smaller values are clipped to after scaling.

  • max_weight – The maximum value, bigger values are clipped to after scaling.

  • as_int – Round to nearest int and return as int type.

Returns

The transformed weight tensor.