File size: 4,039 Bytes
694530d 869843b cd8faa3 869843b 694530d 869843b 27a63b7 869843b 27a63b7 869843b 27a63b7 869843b 27a63b7 869843b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 |
---
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</s>千杯美酒醉风情|<bos>一生只剩诗和酒</s>满腹无关雪与梅</s>|
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：
## Usage
本项目开源在文本生成项目：[textgen](https://github.com/shibing624/textgen)，可支持SongNet模型，通过如下命令调用：
Install package:
```shell
pip install -U textgen
```
```python
import sys
sys.path.append('..')
from textgen.language_modeling import SongNetModel
model = SongNetModel(model_type='songnet', model_name='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:
```shell
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：
```
t5-chinese-couplet
├── pytorch_model.bin
└── vocab.txt
```
### 训练数据集
#### 中文对联数据集
- 数据：[对联github](https://github.com/wb14123/couplet-dataset)、[清洗过的对联github](https://github.com/v-zich/couplet-clean-dataset)
- 相关内容
- [Huggingface](https://huggingface.co/)
- [SongNet paper](https://aclanthology.org/2020.acl-main.68/)
- [textgen](https://github.com/shibing624/textgen)
数据格式：
```text
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](https://github.com/shibing624/textgen/blob/main/examples/language_generation/training_zh_songnet_demo.py)
## Citation
```latex
@software{textgen,
author = {Xu Ming},
title = {textgen: Implementation of Text Generation models},
year = {2022},
url = {https://github.com/shibing624/textgen},
}
```
|