@fchollet Neural GPUs, spatial transformers, attention models would be great
I added a few advanced LSTM examples to the Keras docs: http://keras.io/examples/ What other architectures would you like to see?
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@fchollet, how hard would it be to recreate Google's Inception architecture in Keras? -
@jliemansifry pretty easy (a big Graph model) - Show replies
New conversation -
-
-
@fchollet attention models (or similar)Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@fchollet can you do biLSTM seq tagger, where some input vectors are word-embs, and others are char-level LSTM over the corresponding word? -
- Show replies
New conversation -
-
-
@fchollet a recursive one !Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@fchollet gradient clipping and internal projection in LSTMs (to reach state-of-art architecture in language modelling)Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
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.