Will it be possible to use several backend (if it is not already the case) with Keras? For example, I would like to use a different implementation of the gradient calculus than the one of tensorflow.
We're considering extensive changes to the Keras Preprocessing API to make it compatible with TF Datasets. Please check out the RFC doc here and give us your feedback:https://github.com/keras-team/governance/pull/6 …
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Yes! I really like ImageDataGenerator, but recently switched to TF Datasets for exploration and things like prefetch_to_device.
-
Also mentioned in a threat by
@DynamicWebPaige that TF could natively support data augmentation found in Keras like random rotation and shearinghttps://twitter.com/DynamicWebPaige/status/1144377842576191488?s=19 …
Solicitation for opinions, internet!
Please reply with your wishlist for data preprocessing operations within @TensorFlow / tf.keras. This will be a running thread. Some
s:
- Image preprocessing operations
- Time series handling in feature columns
- Usability enhancements pic.twitter.com/gXsP29z9sQShow this thread
End of conversation
New conversation -
-
-
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.