简介 Brief Introduction


Good at solving NLT tasks, applying the BERT tokenizer, a large-scale Chinese BART.

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 BART 759M 中文-BERT分词器 Chinese-BERTTokenizer

模型信息 Model Information


To obtain a large-scale Chinese BART (around twice as large as BART-large), we use WuDao Corpora (180 GB version) for pre-training. Specifically, we use the fengshen framework in the pre-training phase which cost about 7 days with 8 A100 GPUs. Note that since the BERT tokenizer usually performs better than others for Chinese tasks, we employ it. We have also released our pre-training code: pretrain_randeng_bart.

使用 Usage

模型下载地址 Download Address


加载模型 Loading Models

from transformers import BartForConditionalGeneration, AutoTokenizer, Text2TextGenerationPipeline
import torch

tokenizer=AutoTokenizer.from_pretrained('IDEA-CCNL/Randeng-BART-759M-Chinese-BertTokenizer', use_fast=false)
text = '桂林是著名的[MASK],它有很多[MASK]。'
text2text_generator = Text2TextGenerationPipeline(model, tokenizer)
print(text2text_generator(text, max_length=50, do_sample=False))

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: