moviechats's Introduction

MovieChats

Dataset & Model Download

Preprocessing:

Each dialogue turn will be represented as a sequence, which contains previous user/system turns, dialog act, knowledge(None), and response

[context][user]你觉得血色身材这部电影好看吗？[system]我觉得简直是太精彩了，[user]是吗？[system]就要讲述的故事，一环扣一环，剧情特别的惊心动魄，[user]是不是给人一种特别的紧张感？[system]简直你说的太实在了，[user]是吧，我觉得这部电影算是精品吧，[system]我觉得是精品中的精品，[user]怎么你对这部片儿的评价很高呀？[system]是啊，因为我喜欢看这一种嘛，简直是太好看了，[endofcontext][action][endofaction][knowledge][endofknowledge]
[response]行，那以后我要是有这种电影的话，我还推荐给你[endofresponse]

Usage

unzip ul_model_best.zip unzip pretrain_model.zip unzip movie_data.zip

Requirements

python 2.7+
transformers==2.1.1

Run

./train_ul_best.sh

train the model

python ./train_ul_best.py --epochs 8 --batch_size 64 --pretrained_model ./pretrain_model/pytorch_model.bin

hyper-parameter settings

{
  "initializer_range": 0.02,
  "layer_norm_epsilon": 1e-05,
  "n_ctx": 300,
  "n_embd": 768,
  "n_head": 12,
  "n_layer": 12,
  "n_positions": 300,
  "vocab_size": 13317
}

Reference

@inproceedings{su2020moviechats,
  title={MovieChats: Chat like Humans in a Closed Domain},
  author={Su, Hui and Shen, Xiaoyu and Xiao, Zhou and Zhang, Zheng and Chang, Ernie and Zhang, Cheng and Niu, Cheng and Zhou, Jie},
  booktitle={Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
  pages={6605--6619},
  year={2020}
}

Thanks

GPT2-chitchat(https://github.com/yangjianxin1/GPT2-chitchat) CDial-GPT(https://github.com/thu-coai/CDial-GPT)

moviechats's People

Contributors

Stargazers

Watchers

moviechats's Issues

知识类相关信息会提供出来吗

你好，我看数据集里没有包含知识信息，因为知识类信息的抓取和对齐会比较麻烦，特别是评论方面的对齐，不知道后续会不会提供出来呢，谢谢~

pretrain model has some problems

I have downloaded the chitchat_pretrain_model from the google drive. while running the GPT2LMHeadModel.from_pretrained, raise that "RuntimeError: unexpected EOF, expected 5249182 more bytes. The file might be corrupted."

测试疑问

作者你好，很高兴你们提出了一个end2end的框架来进行知识驱动的对话生成，并提供了更自然更大的数据集。但是对文中模型那块有点不理解。Movie Tracker、Intent Prediction、Knowledge Retrieval、Text Generation这四个子任务是串行运行的吗？先预测出movie tracker然后填到相应的位置进行intent的预测？因为不太明白这一点，所以对代码进行阅读，但是测试代码中对于Movie Tracker、Intent Prediction、Knowledge Retrieval这部分内容好像没有涉及？期待您的回复，万分感谢！

Recommend Projects