@fchollet I was thinking the same... But makes sense since this is a feedforward convnet with max-margin cost which is built in in Torch.
Surprising opinions; many people would advocate the opposite (Keras for experimentation, Torch for production).https://twitter.com/edersantana/status/705929356095451136 …
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@fchollet they'd have to use a graph with the recent siamese layer to train that in Keras. Maybe that is also why...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.