represents a net layer that applies random image transformations to produce images of height h and width w.

Details and Options


open allclose all

Basic Examples  (2)

Create an ImageAugmentationLayer whose output dimensions are {n,80,80}:

Create an ImageAugmentationLayer that takes an image of size 128×128 and returns an image crop of size 80×80:

Apply the layer to an image, obtaining the center crop:

Apply the layer to an image, specifying that training behavior be used:

The layer threads across a batch of examples:

Options  (1)

"ReflectionProbabilities"  (1)

Create an ImageAugmentationLayer that randomly reflects an image about the vertical axis with probability 0.9 during training:

Apply the layer to an image, using training behavior:

Possible Issues  (1)

Currently, any randomness invoked by NetEvaluationMode->"Train" is not affected by SeedRandom and BlockRandom:

Introduced in 2017