Skip to content

Code used by the paper "What is the Role of Recurrent Neural Networks (RNNs) in an Image Caption Generator?".

License

Notifications You must be signed in to change notification settings

mtanti/rnn-role

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

rnn-role

Code used by the paper "What is the Role of Recurrent Neural Networks (RNNs) in an Image Caption Generator?". https://arxiv.org/abs/1703.09137

This paper investigates whether the RNN should be viewed as a sequence generator or as a sequence encoder. It does so by comparing the performance of a caption generator when the image features are included in the RNN or outside of the RNN. Paper will appear at the INLG 2017 conference.

Works on both python 2 and python 3 (except for the MSCOCO evaluation toolkit which requires python 2) and uses TensorFlow 1.0.

Python dependencies (install all with pip):

  • tensorflow
  • future

About

Code used by the paper "What is the Role of Recurrent Neural Networks (RNNs) in an Image Caption Generator?".

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages