Shashi Narayan

For training you only need (1). (2) is used to estimate ROUGE scores. (1) and (3) is used during decoding. What error do you get after the first epoch?

It should point to 3) doc_sentence_directory directory.

Please check: https://github.com/shashiongithub/sidenet/issues/2

The Idea presented here could be easily applied to word level extraction. I am not sure if you want to keep the convolutional sentence decoder (it will depend on the...

Hi Allen, Sounds very interesting. However, I have moved from Edinburgh recently. I would be very happy to discuss with you in depth about any implementation issues.

Sure no problem!

They can be found in "Dataset with sideinfo: http://kinloch.inf.ed.ac.uk/public/cnn-dm-sideinfo-data.zip". But, I have also updated the Readme page to directly download those preprocessed files.

Ohh thanks, I have uploaded them as well. Shashi

it is not removed. It is available as a tar file.

Sorry, I misunderstood. You meant in Version V-0.1. We removed the split file because there was a bug in our split. We recommend you to use Roee's split for this....