简介 Brief Introduction


Good at solving text summarization tasks, Chinese PAGASUS-large.

模型分类 Model Taxonomy

需求 Demand 任务 Task 系列 Series 模型 Model 参数 Parameter 额外 Extra
通用 General 自然语言转换 NLT 燃灯 Randeng PEFASUS 523M 中文 Chinese

模型信息 Model Information

参考论文:PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization

为了解决中文的自动摘要任务,我们遵循PEGASUS的设计来训练中文的版本。我们使用了悟道语料库(180G版本)作为预训练数据集。此外,考虑到中文sentence piece不稳定,我们在Randeng-PEGASUS中同时使用了结巴分词和BERT分词器。我们也提供base的版本:IDEA-CCNL/Randeng-Pegasus-238M-Chinese。以及,我们也提供了在中文摘要数据集上微调的版本:Randeng-Pegasus-523M-Summary-Chinese

To solve Chinese abstractive summarization tasks, we follow the PEGASUS guidelines. We employ a version of WuDao Corpora (180 GB version) as a pre-training dataset. In addition, considering that the Chinese sentence chunk is unstable, we utilize jieba and BERT tokenizer in our Randeng-PEGASUS. We also provide a base size version, available with IDEA-CCNL/Randeng-Pegasus-238M-Chinese. And, we also provide a version after fine-tuning on Chinese text summarization datasets: Randeng-Pegasus-523M-Summary-Chinese.

使用 Usage

模型下载地址 Download Address


加载模型 Loading Models

from transformers import PegasusForConditionalGeneration
# Need to download tokenizers_pegasus.py and other Python script from Fengshenbang-LM github repo in advance,
# or you can download tokenizers_pegasus.py and data_utils.py in https://huggingface.co/IDEA-CCNL/Randeng_Pegasus_523M/tree/main
# Strongly recommend you git clone the Fengshenbang-LM repo:
# 1. git clone https://github.com/IDEA-CCNL/Fengshenbang-LM
# 2. cd Fengshenbang-LM/fengshen/examples/pegasus/
# and then you will see the tokenizers_pegasus.py and data_utils.py which are needed by pegasus model
from tokenizers_pegasus import PegasusTokenizer

model = PegasusForConditionalGeneration.from_pretrained("IDEA-CCNL/Randeng-Pegasus-523M-Chinese")
tokenizer = PegasusTokenizer.from_pretrained("IDEA-CCNL/Randeng-Pegasus-523M-Chinese")

text = "据微信公众号“界面”报道,4日上午10点左右,中国发改委反垄断调查小组突击查访奔驰上海办事处,调取数据材料,并对多名奔驰高管进行了约谈。截止昨日晚9点,包括北京梅赛德斯-奔驰销售服务有限公司东区总经理在内的多名管理人员仍留在上海办公室内"
inputs = tokenizer(text, max_length=1024, return_tensors="pt")

# Generate Summary
summary_ids = model.generate(inputs["input_ids"])
tokenizer.batch_decode(summary_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]

# model Output: 截止昨日晚9点,包括北京梅赛德斯-奔驰销售服务有限公司东区总经理在内的多名管理人员仍留在上海办公室内

引用 Citation


If you are using the resource for your work, please cite the our paper:

  author    = {Junjie Wang and Yuxiang Zhang and Lin Zhang and Ping Yang and Xinyu Gao and Ziwei Wu and Xiaoqun Dong and Junqing He and Jianheng Zhuo and Qi Yang and Yongfeng Huang and Xiayu Li and Yanghan Wu and Junyu Lu and Xinyu Zhu and Weifeng Chen and Ting Han and Kunhao Pan and Rui Wang and Hao Wang and Xiaojun Wu and Zhongshen Zeng and Chongpei Chen and Ruyi Gan and Jiaxing Zhang},
  title     = {Fengshenbang 1.0: Being the Foundation of Chinese Cognitive Intelligence},
  journal   = {CoRR},
  volume    = {abs/2209.02970},
  year      = {2022}


You can also cite our website: