torchsig.transforms.dataset_transforms.CutOut

class torchsig.transforms.dataset_transforms.CutOut(duration=(0.01, 0.2), cut_type: List[str] = ['zeros', 'ones', 'low_noise', 'avg_noise', 'high_noise'], **kwargs)

Bases: DatasetTransform

Applies the CutOut transform operation in the time domain. The cut_dur input specifies how long the cut region should be, and the cut_fill input specifies what the cut region should be filled in with. Options for the cut type include: zeros, ones, low_noise, avg_noise, and high_noise. Zeros fills in the region with zeros; ones fills in the region with 1+1j samples; low_noise fills in the region with noise with -100dB power; avg_noise adds noise at power average of input data, effectively slicing/removing existing signals in the most RF realistic way of the options; and high_noise adds noise with 40dB power. If a list of multiple options are passed in, they are randomly sampled from.

This transform is loosely based on “Improved Regularization of Convolutional Neural Networks with Cutout”.

duration

cut_dur sets the duration of the region to cut out * If float, cut_dur is fixed at the value provided. * If list, cut_dur is any element in the list. * If tuple, cut_dur is in range of (tuple[0], tuple[1]).

Type:

float, list, tuple

duration_distribution

Random draw from duration distribution.

Type:

Callable[[], float]

cut_type

cut_fill sets the type of data to fill in the cut region with from the options: zeros, ones, low_noise, avg_noise, and high_noise * If list, cut_fill is any element in the list. * If str, cut_fill is fixed at the method provided.

Type:

float, list, tuple

cut_type_distribution

Random draw from cut_type distribution.

Type:

Callable[[], str]

Methods

add_parent	Add parent Seedable object and set up RNGs accordingly
get_distribution
get_second_seed	Gets second seed, usually used to seed both torch and numpy generators with slightly different seeds
seed	Seed number generators with given seed.
setup_rngs	Initialize torch and numpy number generators, and update its children.
update	Updates bookkeeping to transforms in DatasetSignal's SignalMetadata and checks signal valididty.
update_from_parent	Update numpy and torch number generators with parent seed

__init__(duration=(0.01, 0.2), cut_type: List[str] = ['zeros', 'ones', 'low_noise', 'avg_noise', 'high_noise'], **kwargs)

Transform initialization as Seedable.

__call__(signal: DatasetSignal) → DatasetSignal

Performs transforms.

Parameters:

signal (DatasetSignal) – DatasetSignal to be transformed.

Raises:

NotImplementedError – Inherited classes must override this method.

Returns:

Transformed DatasetSignal.

Return type:

DatasetSignal

__repr__() → str

Transform string representation. Should be able to recreate class from this string.

Returns:

Transform representation.

Return type:

str

__str__() → str

Return str(self).

add_parent(parent) → None

Add parent Seedable object and set up RNGs accordingly

get_second_seed(seed: int) → int

Gets second seed, usually used to seed both torch and numpy generators with slightly different seeds

Parameters:

seed (int) – Seed to use.

Returns:

New seed.

Return type:

int

seed(seed: int) → None

Seed number generators with given seed.

Parameters:

seed (int) – Seed to use.

setup_rngs() → None

Initialize torch and numpy number generators, and update its children.

update(signal: DatasetSignal) → None

Updates bookkeeping to transforms in DatasetSignal’s SignalMetadata and checks signal valididty. Inherited classes should always call self.update() after performing transform operation (inside __call__).

Parameters:

signal (DatasetSignal) – transformed DatasetSignal.

update_from_parent() → None

Update numpy and torch number generators with parent seed