Augmenting images?

General Discussion
, ,
#1

I apologize. I am new to tensorflow. My question is most likely obvious (but not to me). I have seen a lot of image augmentation that uses tf.keras.preprocessing.image.ImageGenerator(). In a tensorflow hub example, the author used:

do_data_augmentation = False
if do_data_augmentation:
  preprocessing_model.add(
      tf.keras.layers.RandomRotation(40))
  preprocessing_model.add(
      tf.keras.layers.RandomTranslation(0, 0.2))
  preprocessing_model.add(
...

The other notes the ImageDataGenerator() method is “old”. Why is it old? Why is the new method better? Thank you.

#3

@Margaret_Johnson,

Welcome to the Tensorflow Forum!

tf.keras.preprocessing.image.ImageDataGenerator is obsolete, which is not recommended in the new code. Prefer loading images with tf.keras.utils.image_dataset_from_directory and transforming the output tf.data.Dataset with preprocessing layers. For more information, see the tutorials for loading images and augmenting images as well as the preprocessing layer guide.

Thank you!