Kyunghyun Cho

@AjayTalati I don't fully understand what you mean by "no internal state tracking". If there's no internal state, by definition is not a recurrent neural network (hence, no LSTM.) What...

I think I understand what you meant by 'internal state tracking'. So, essentially, you want to do language modelling (next symbol prediction) with RNN (but not necessarily with LSTM). Now,...

@nicholas-leonard Thanks! And, for proper credit assignment, thanks to @carriepl

Unfortunately, I don't have time to implement STORN at the moment. On Fri, Mar 13, 2015 at 10:54 AM, Ajay Talati [email protected] wrote: > So coding the LSTM variational autoencoder...

Ajay, feel free to contact me or @nouis, if you run into any issue implementing LSTM with Theano. On Fri, Mar 13, 2015 at 6:42 PM, Ajay Talati [email protected] wrote:...

since it's unsupervised, we can't expect exact match to be high. we should probably use a metric that measures whether the answer span is included in a highlighted sentence.

since we don't want to break re-ranking too much, how about "chronologically sort top-50"?

@leo-zhou @jnhwkim is correct, except that `feat` is saved as a sparse matrix of shape `[N, 14 * 14 * 512]`. In the dictionary (dictionary.pkl), 0 and 1 are reserved...

It's likely because you didn't use "protocol=cPickle.HIGHEST_PROTOCOL" as an argument with cPickle.dump. - K On Wed, Nov 4, 2015 at 10:54 AM, Yizhou Hu [email protected] wrote: > @kelvinxu https://github.com/kelvinxu The...