1

I would like to use tensorflows random flip to train the generator part of a gan network only. For that i would like a way to either flip the image randomly at the begining of the model and the end. I don't want to augment the dataset because right now it fits my RAM quit nicely.

NameVergessen
  • 598
  • 8
  • 26
  • Hi! You can just put ImageDataGenerator which includes Data augmentation features like Flipping the image, blurring, rotation etc in the model creation or loading the dataset itself . Ref [1](https://www.tensorflow.org/tutorials/images/data_augmentation#two_options_to_use_the_keras_preprocessing_layers), [2](https://www.tensorflow.org/api_docs/python/tf/keras/preprocessing/image/ImageDataGenerator) –  Jul 14 '22 at 05:47
  • Will this work for Imput and Output Image? I only ever see the capability to flip the input image. – NameVergessen Jul 14 '22 at 09:42
  • Yes. You can augment in output images too . Just use the augmentation feature from layers instead of imageGenerators(you can use it in input). Thank you! –  Jul 14 '22 at 10:46

0 Answers0