Simple Keras SNLI baseline where sum(GloVe) beats several deep learning baselines https://github.com/Smerity/keras_snli …pic.twitter.com/50agaKGj5O
-
tldr Model is essentially
@sleepinyourhat's 2015 with tweaks imported from latter papers.@fchollet, get GloVe in model zoo ;)
problem with GloVe in zoo is that the matrix should be filtered for the voc at hand e.g. https://blog.keras.io/using-pre-trained-word-embeddings-in-a-keras-model.html …
-
-
if you see a way to automate / simplify the process then sure, we could do it
-
maybe get_glove_embedding(dim, word_index)
End of conversation
New conversation -
-
-
Gah, good point, +1. Cache that stores filtered GloVe based upon vocab (Vectorizer) as key? Sounds too complicated. Hmm ...
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.