The nice thing about Keras running on TensorFlow is that I can do neural style transfer on CPU, kinda fast. Would be unthinkable with Theano
-
@fchollet so one can train a model using the Theano backend and after that doing learning transfer using the TensorFlow backend?
@Miguel_Molero yes. You can load a Theano-trained model with TF and conversely.
-
-
@fchollet amazing powerful tool you've created, thanks for sharing. Indeed I already loved Keras just with Theano and now more with thisThanks. 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.