tfm.vision.augment.RandAugment

Applies the RandAugment policy to images.

Inherits From: ImageAugment

tfm.vision.augment.RandAugment(
    num_layers: int = 2,
    magnitude: float = 10.0,
    cutout_const: float = 40.0,
    translate_const: float = 100.0,
    magnitude_std: float = 0.0,
    prob_to_apply: Optional[float] = None,
    exclude_ops: Optional[List[str]] = None
)

RandAugment is from the paper https://arxiv.org/abs/1909.13719

num_layers Integer, the number of augmentation transformations to apply sequentially to an image. Represented as (N) in the paper. Usually best values will be in the range [1, 3].
magnitude Integer, shared magnitude across all augmentation operations. Represented as (M) in the paper. Usually best values are in the range [5, 10].
cutout_const multiplier for applying cutout.
translate_const multiplier for applying translation.
magnitude_std randomness of the severity as proposed by the authors of the timm library.
prob_to_apply The probability to apply the selected augmentation at each layer.
exclude_ops exclude selected operations.

Methods

build_for_detection

View source

@classmethod
build_for_detection(
    num_layers: int = 2,
    magnitude: float = 10.0,
    cutout_const: float = 40.0,
    translate_const: float = 100.0,
    magnitude_std: float = 0.0,
    prob_to_apply: Optional[float] = None,
    exclude_ops: Optional[List[str]] = None
)

Builds a RandAugment that modifies bboxes for geometric transforms.

distort

View source

distort(
    image: tf.Tensor
) -> tf.Tensor

See base class.

distort_with_boxes

View source

distort_with_boxes(
    image: tf.Tensor, bboxes: tf.Tensor
) -> Tuple[tf.Tensor, tf.Tensor]

See base class.