MPQG

MPQG copied to clipboard

how to evaluate the output

1

I have decoded and get the test.sota.tok file, how can i get the evaluation of it?

SideOfTheRoad

Statistics of released dataset splits

5

Hey @freesunshine0316, Would you please confirm the stats of train/dev/test in your experiment? With the released data, I am getting: ~ 75500/17934/11805 Is this correct? Thanks

deepaknlp

Data logic

3

Hi, I'm trying to understand how text1, text2 and text3 are being derived in your experiment as there is no understanding of that in the paper. What is your logic...

HarbingerOfHell

Extending Vocabulary

1

How can we extend the vocabulary ? Currently when we generate questions we get a lot of unk tokens.

avimeens

Unable to achieve the score released by paper

1

I execute the source code and the parameters have not changed, but the score I got was only 11(BLEU-4). The score released by paper is 13.98(BLEU-4). I want to ask...

leechehao

encoder state in `matching_encoder_utils.py`

4

I read the paper "Leveraging Context Information for Natural Question Generation". Section 2.2 says: > Each encoder state hj is the concatenation of two bi-directional LSTM states (Section 3.2 also...

pjlintw