kaira.models.binary.soft_bit_thresholding.SoftBitEnsembleThresholder

Inheritance diagram for SoftBitEnsembleThresholder

class kaira.models.binary.soft_bit_thresholding.SoftBitEnsembleThresholder(thresholders: List[SoftBitThresholder], voting: str = 'majority', weights: Tensor | List[float] | None = None, *args: Any, **kwargs: Any)

Bases: SoftBitThresholder

Ensemble thresholder that combines decisions from multiple thresholders.

This thresholder aggregates the outputs of multiple thresholding approaches using various voting schemes to produce a more robust decision. This can be particularly effective in challenging noise conditions or when the optimal thresholding strategy is unclear.

Example

With thresholders=[FixedThresholder(), AdaptiveThresholder()] and voting=’majority’: The output will be 1.0 only if both thresholders output 1.0.

Methods

__init__	Initialize the ensemble thresholder.
forward	Apply ensemble thresholding to convert soft bit values to hard decisions.
to	Move the model to the specified device.

__init__(thresholders: List[SoftBitThresholder], voting: str = 'majority', weights: Tensor | List[float] | None = None, *args: Any, **kwargs: Any)

Initialize the ensemble thresholder.

Parameters:

• thresholders – List of thresholders to combine.

• voting – Voting strategy to use: ‘majority’, ‘weighted’, ‘any’, or ‘all’.

• weights – Optional weights for each thresholder (used with ‘weighted’ voting).

• *args – Variable positional arguments passed to the base class.

• **kwargs – Variable keyword arguments passed to the base class.

forward(x: Tensor, *args: Any, **kwargs: Any) → Tensor

Apply ensemble thresholding to convert soft bit values to hard decisions.

Parameters:

• x – Input tensor of soft bit values.

• *args – Additional positional arguments passed to thresholders.

• **kwargs – Additional keyword arguments passed to thresholders.

Returns:

Tensor of hard bit decisions (0.0 or 1.0).

to(device: str | device, *args, **kwargs) → SoftBitThresholder

Move the model to the specified device.

Parameters:

• device – The device to move the model to.

• *args – Additional positional arguments for nn.Module.to().

• **kwargs – Additional keyword arguments for nn.Module.to().

Returns:

Self for method chaining.