cvnets.neural_augmentor.utils package

Submodules

cvnets.neural_augmentor.utils.neural_aug_utils module

class cvnets.neural_augmentor.utils.neural_aug_utils.Clip(min_val: float, max_val: float, hard_clip: bool | None = False, *args, **kwargs)

Bases: Module

__init__(min_val: float, max_val: float, hard_clip: bool | None = False, *args, **kwargs) → None

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(x: Any) → Any

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class cvnets.neural_augmentor.utils.neural_aug_utils.Identity(*args, **kwargs)

Bases: Module

__init__(*args, **kwargs)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(x: Any) → Any

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class cvnets.neural_augmentor.utils.neural_aug_utils.FixedSampler(value: float, clip_fn: Module | None = Identity(), *args, **kwargs)

Bases: Module

__init__(value: float, clip_fn: Module | None = Identity(), *args, **kwargs)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(sample_shape=(), data_type=torch.float32, device=device(type='cpu')) → Tensor

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class cvnets.neural_augmentor.utils.neural_aug_utils.UniformSampler(low: float, high: float, min_fn: Module | None = Identity(), max_fn: Module | None = Identity(), *args, **kwargs)

Bases: Module

__init__(low: float, high: float, min_fn: Module | None = Identity(), max_fn: Module | None = Identity(), *args, **kwargs)

Initializes internal Module state, shared by both nn.Module and ScriptModule.

forward(sample_shape=(), data_type=torch.float32, device=device(type='cpu')) → Tensor

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

property high

property low

cvnets.neural_augmentor.utils.neural_aug_utils.random_noise(x: Tensor, variance: Tensor, *args, **kwargs) → Tensor

Apply random noise sampled.

cvnets.neural_augmentor.utils.neural_aug_utils.random_contrast(x: Tensor, magnitude: Tensor, *args, **kwargs) → Tensor

cvnets.neural_augmentor.utils.neural_aug_utils.random_brightness(x: Tensor, magnitude: Tensor, *args, **kwargs) → Tensor

Brightness function.

cvnets.neural_augmentor.utils.neural_aug_utils.identity(x: Tensor, *args, **kwargs) → Tensor

Identity function

Module contents