AugMix layer


AugMix class

    chain_depth=[1, 3],

Performs the AugMix data augmentation technique.

AugMix aims to produce images with variety while preserving the image semantics and local statistics. During the augmentation process, each image is augmented num_chains different ways, each way consisting of chain_depth augmentations. Augmentations are sampled from the list: translation, shearing, rotation, posterization, histogram equalization, solarization and auto contrast. The results of each chain are then mixed together with the original image based on random samples from a Dirichlet distribution.


  • value_range: the range of values the incoming images will have. Represented as a two number tuple written (low, high). This is typically either (0, 1) or (0, 255) depending on how your preprocessing pipeline is setup.
  • severity: A tuple of two floats, a single float or a keras_cv.FactorSampler. A value is sampled from the provided range. If a float is passed, the range is interpreted as (0, severity). This value represents the level of strength of augmentations and is in the range [0, 1]. Defaults to 0.3.
  • num_chains: an integer representing the number of different chains to be mixed. Defaults to 3.
  • chain_depth: an integer or range representing the number of transformations in the chains. Defaults to [1,3].
  • alpha: a float value used as the probability coefficients for the Beta and Dirichlet distributions. Defaults to 1.0.
  • seed: Integer. Used to create a random seed.


Sample Usage:

(images, labels), _ = tf.keras.datasets.cifar10.load_data()
augmix = keras_cv.layers.AugMix([0, 255])
augmented_images = augmix(images[:100])