The killer app is to implement something like a GAN training step in 20 lines and call `fit()` on it -- no further work needed to train on 8 GPUs or on TPUpic.twitter.com/Wyu2lBaAaJ
Show this thread
Keras Tweetorial: this is how you implement custom training logic that you can use with `fit()` (shown here: just a simple supervised training loop). This workflow enables you to instantly leverage callbacks, distributed tuning, TensorBoard, etc. with your own training logicpic.twitter.com/LtEx20SMj5
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Can we have tutorial on writing scalars to tensorboard with this kind of set up ??
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
How are you able to just call fit()? Is train_step() being overrided?
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.