► Keras 3 API documentation / Layers API / Preprocessing layers / Image augmentation layers / RandAugment layer

RandAugment layer

[source]

RandAugment class

keras.layers.RandAugment(
    value_range=(0, 255),
    num_ops=2,
    factor=0.5,
    interpolation="bilinear",
    seed=None,
    data_format=None,
    **kwargs
)

RandAugment performs the Rand Augment operation on input images.

This layer can be thought of as an all-in-one image augmentation layer. The policy implemented by this layer has been benchmarked extensively and is effective on a wide variety of datasets.

References

Arguments

  • value_range: The range of values the input image can take. Default is (0, 255). Typically, this would be (0, 1) for normalized images or (0, 255) for raw images.
  • num_ops: The number of augmentation operations to apply sequentially to each image. Default is 2.
  • factor: The strength of the augmentation as a normalized value between 0 and 1. Default is 0.5.
  • interpolation: The interpolation method to use for resizing operations. Options include nearest, bilinear. Default is bilinear.
  • seed: Integer. Used to create a random seed.
RandAugment layer
RandAugment class