Paper of the source codes released:

This repository contains the source code and datasets for the model Multi-hop Selector Network.

Dependencies

Python 3.x
Pytorch 1.1.0

Datasets and Trained Models

Your can download the processed datasets and the checkpoints of trained models for reproduce the experimental results in the paper by the following url:
https://drive.google.com/drive/folders/1pJKIppcbjuTZxbTc8ye5mfnC2ygR2xTo?usp=sharing

Unzip the dataset.rar file to the folder of dataset and unzip the checkpoint.rar file to the folder of checkpoint.
The log/ directory already contains the training and evaluation logs of each dataset.

Reproduce the experimental results by the pre-trained model

cd ./Dialogue/
python ./run.py --task "ubuntu"
python ./run.py --task "douban"
python ./run.py --task "alime"

Train a new model

cd ./Dialogue/
python ./run.py --task "ubuntu" --is_training True
python ./run.py --task "douban" --is_training True
python ./run.py --task "alime" --is_training True

The training process is recorded in log/[ubuntu|douban|alime].msn.log file.

Citation

If you find this code useful in your research, please cite our paper:

@inproceedings{yuan2019multi,
  title={Multi-hop Selector Network for Multi-turn Response Selection in Retrieval-based Chatbots},
  author={Yuan, Chunyuan and Zhou, Wei and Li, Mingming and Lv, Shangwen and Zhu, Fuqing and Han, Jizhong and Hu, Songlin},
  booktitle={Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)},
  pages={111--120},
  year={2019}
}

chunyuanY / Dialogue