Discourse-Aware Response Generation via Learning to Recover and Rank Utterances

Aug 24, 2021 1 min read

DialogBERT

This is a PyTorch implementation of the DialogBERT model described in
DialogBERT: Neural Response Generation via Hierarchical BERT with Distributed Utterance Order Ranking.

Prerequisites

Python 3.6
PyTorch

Install packages of the requirements.txt file.

Usage

Train model by
```
  python main.py
```

The logs and temporary results will be printed to stdout and saved in the ./output path.

Run test by
```
python main.py --do_test --reload_from XXXXX
```
where XXXXX specifies the iteration number of the optimal checkpoint.

References

If you use any source code included in this toolkit in your work, please cite the following paper:

@inproceedings{gu2021dialogbert,
      title={Dialog{BERT}: Discourse-Aware Response Generation via Learning to Recover and Rank Utterances},
      author={Gu, Xiaodong and Yoo, Kang Min and Ha, Jung-Woo},
      journal={In Proceedings of the 35th AAAI Conference on Artificial Intelligence (AAAI 2021)},
      year={2021}
}

GitHub

https://github.com/guxd/DialogBERT

John was the first writer to have joined pythonawesome.com. He has since then inculcated very effective writing and reviewing culture at pythonawesome which rivals have found impossible to imitate.