shibing624's picture
Update README.md
8db4031
metadata
language:
  - zh
tags:
  - SongNet
  - pytorch
  - zh
  - Text2Text-Generation
license: apache-2.0
widget:
  - text: 丹枫江冷人初去

SongNet for Chinese Couplet(songnet-base-chinese-couplet) Model

SongNet中文对联仿写模型

songnet-base-chinese-couplet evaluate couplet test data:

The overall performance of SongNet on couplet test:

input_text predict
一句相思吟岁月,千杯美酒醉风情 一生只剩诗和酒,满腹无关雪与梅

在Couplet测试集上生成结果满足字数相同、词性对齐、词面对齐、形似要求,针对性的SongNet网络结构,在语义对仗工整和平仄合律上的效果明显优于T5和GPT2等模型。

SongNet的网络结构:

arch

Usage

本项目开源在文本生成项目:textgen,可支持SongNet模型,通过如下命令调用:

Install package:

pip install -U textgen
import sys

sys.path.append('..')
from textgen.language_modeling import SongNetModel


model = SongNetModel(model_type='songnet', model_name='shibing624/songnet-base-chinese-couplet')
sentences = [
    "严蕊<s1>如梦令<s2>道是梨花不是。</s>道是杏花不是。</s>白白与红红,别是东风情味。</s>曾记。</s>曾记。</s>人在武陵微醉。",
    "<s1><s2>一句相思吟岁月</s>千杯美酒醉风情",
    "<s1><s2>几树梅花数竿竹</s>一潭秋水半屏山"
    "<s1><s2>未舍东江开口咏</s>且施妙手点睛来",
    "<s1><s2>一去二三里</s>烟村四五家",
]
print("inputs:", sentences)
print("outputs:", model.generate(sentences))
sentences = [
    "<s1><s2>一句____月</s>千杯美酒__情",
    "<s1><s2>一去二三里</s>烟村__家</s>亭台__座</s>八__枝花",
]
print("inputs:", sentences)
print("outputs:", model.fill_mask(sentences))

output:

inputs: ['严蕊<s1>如梦令<s2>道是梨花不是。</s>道是杏花不是。</s>白白与红红,别是东风情味。</s>曾记。</s>曾记。</s>人在武陵微醉。', '<s1><s2>一句相思吟岁月</s>千杯美酒醉风情', '<s1><s2>几树梅花数竿竹</s>一潭秋水半屏山<s1><s2>未舍东江开口咏</s>且施妙手点睛来', '<s1><s2>一去二三里</s>烟村四五家']
outputs: ['<bos>盛世欣开新气象</s>春联喜绘大文章</s>春天铺锦笺,宏图更写好山山</s>新篇章</s>新篇章</s>神州高唱好年华</s>', '<bos>一曲琴音添雅韵</s>几回酒醉解愁思</s>', '<bos>三分天下隆中对</s>四面八方九派江山笔底留</s>', '<bos>春深花已老</s>夜静露方浓</s>']
inputs: ['<s1><s2>一句____月</s>千杯美酒__情', '<s1><s2>一去二三里</s>烟村__家</s>亭台__座</s>八__枝花']
outputs: ['<bos>一句佳诗吟盛月</s>千杯美酒祝春情</s>', '<bos>一去二三里</s>烟村百二家</s>亭台十二座</s>八里一枝花</s>']

模型文件组成:

songnet-base-chinese-couplet
    ├── pytorch_model.bin
    └── vocab.txt

训练数据集

中文对联数据集

数据格式:

head -n 1 couplet_files/couplet/train/in.txt
晚 风 摇 树 树 还 挺 

head -n 1 couplet_files/couplet/train/out.txt
晨 露 润 花 花 更 红 

如果需要训练SongNet模型,请参考https://github.com/shibing624/textgen/blob/main/examples/language_generation/training_zh_songnet_demo.py

Citation

@software{textgen,
  author = {Xu Ming},
  title = {textgen: Implementation of Text Generation models},
  year = {2022},
  url = {https://github.com/shibing624/textgen},
}