Could either of you please briefly explain why it is much faster to apply the augmentation after batching? I've been doing it the other way around all this time...
New code example on http://keras.io , from @RisingSayak: using RandAugment from the imgaug library in a http://tf.data pipeline, to train more robust image classification models.https://keras.io/examples/vision/randaugment/ …
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.