为了促进中文自然语言处理研究的发展，本项目提供了 CPM-LM (2.6B) 模型的文本生成代码，可用于文本生成的本地测试，并以此为基础进一步研究零次学习/少次学习等场景。[项目首页] [模型下载]
首先安装pytorch等基础依赖，再安装APEX以支持fp16：
pip install -r requirements.txt
git clone https://github.com/NVIDIA/apex
cd apex
pip install -v --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./
模型下载后文件夹的目录结构需设置如下：
.
├── 80000
│ ├── mp_rank_00_model_states.pt
│ └── mp_rank_01_model_states.pt
└── latest_checkpointed_iteration.txt
为保证下载文件的正确性，文件的checksum如下：
SHA1
7a8461098dae374a836f55bf2002bf21cba05964 80000/mp_rank_00_model_states.pt
6ab97bd06979c1707080b33cfddff4f08c9af675 80000/mp_rank_01_model_states.pt
MD5
ffb39d8e477aedf9f8714cc467f34eae 80000/mp_rank_00_model_states.pt
8af02c1aa85e76a430d45f3ad073edc5 80000/mp_rank_01_model_states.pt
提供了命令行交互式生成：
bash scripts/generate_text.sh /path/to/CPM
运行该脚本需要两块GPU，每张卡的GPU内存占用约为7GB。该项目主要基于 Megatron-LM 进行修改。
@article{cpm-v1,
title={CPM: A Large-scale Generative Chinese Pre-trained Language Model},
author={Zhang, Zhengyan and Han, Xu, and Zhou, Hao, and Ke, Pei, and Gu, Yuxian and Ye, Deming and Qin, Yujia and Su, Yusheng and Ji, Haozhe and Guan, Jian and Qi, Fanchao and Wang, Xiaozhi and Zheng, Yanan and Cao, Jiannan and Zeng, Guoyang and Cao, Huanqi and Chen, Shengqi and Li, Daixuan and Sun, Zhenbo and Liu, Zhiyuan and Huang, Minlie and Han, Wentao and Tang, Jie and Li, Juanzi and Sun, Maosong},
year={2020}
}