therne / dmn-tensorflow

Dynamic Memory Networks (https://arxiv.org/abs/1603.01417) in Tensorflow
240 stars 82 forks source link

Input module in the dmn+ #10

Closed AIari closed 7 years ago

AIari commented 7 years ago

In the input module how to embed the information that coming from the context? In some papers it has mentioned

The will concatenate all the words in the context and add an EOS at the end of each sentence and feed in through r RNN with GRU units. Then take the hidden states at each time step.

I went through the code it's bit different here. What is actually happening ? Each sentence separately feed through a RNN ?