kaira.models.binary.soft_bit_thresholding.AdaptiveThresholder

Inheritance diagram for AdaptiveThresholder

class kaira.models.binary.soft_bit_thresholding.AdaptiveThresholder(method: str = 'mean', scale_factor: float = 1.0, input_type: InputType = InputType.PROBABILITY, *args: Any, **kwargs: Any)

Bases: SoftBitThresholder

Adaptive thresholder for soft bit values.

Adjusts the threshold based on the statistics of the input signal. This can be useful in varying channel conditions where a fixed threshold may not be optimal.

Supports different adaptive threshold methods: - ‘mean’: Uses the mean of the input as threshold - ‘median’: Uses the median of the input as threshold - ‘otsu’: Uses Otsu’s method for optimal bimodal threshold

Methods

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

__init__(method: str = 'mean', scale_factor: float = 1.0, input_type: InputType = InputType.PROBABILITY, *args: Any, **kwargs: Any)

Initialize the adaptive thresholder.

Parameters:

• method – Method to use for adaptive thresholding (‘mean’, ‘median’, ‘otsu’).

• scale_factor – Factor to scale the computed threshold.

• input_type – Type of soft input (‘prob’ or ‘llr’).

• *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 adaptive thresholding to convert soft bit values to hard decisions.

Parameters:

• x – Input tensor of soft bit values.

• *args – Additional positional arguments (unused).

• **kwargs – Additional keyword arguments (unused).

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.