---
title: Erlangshen-MegatronBert-1.3B-NLI
canonical_url: "https://www.modelscope.cn/models/Fengshenbang/Erlangshen-MegatronBert-1.3B-NLI"
md_url: "https://www.modelscope.cn/models/Fengshenbang/Erlangshen-MegatronBert-1.3B-NLI.md"
repository: Fengshenbang/Erlangshen-MegatronBert-1.3B-NLI
chinese_name: "二郎神-MegatronBert-1.3B-NLI 自然语言理解"
last_updated: 2023-02-27
license: "Apache License 2.0"
pipeline_tag: fill-mask
tasks:
  - fill-mask
model_type:
  - megatron-bert
architectures:
  - MegatronBertForSequenceClassification
library_name:
  - pytorch
frameworks:
  - pytorch
language:
  - zh
downloads: 309
stars: 0
tags:
  - bert
  - NLU
  - NLI
---

# Erlangshen-MegatronBert-1.3B-NLI

> Erlangshen-MegatronBert-1.3B-NLI - Fengshenbang 在 ModelScope 开源的模型。二郎神-MegatronBert-1.3B-NLI 自然语言理解

Fengshenbang/Erlangshen-MegatronBert-1.3B-NLI 是 ModelScope 魔搭社区上的fill-mask模型，采用 Apache License 2.0 许可。

- **Repository**: Fengshenbang/Erlangshen-MegatronBert-1.3B-NLI
- **License**: Apache License 2.0
- **Tasks**: fill-mask
- **Tags**: bert, NLU, NLI
- **Downloads**: 309
- **Stars**: 0
- **Last updated**: 2023-02-27

Source: https://www.modelscope.cn/models/Fengshenbang/Erlangshen-MegatronBert-1.3B-NLI

---

# 二郎神-MegatronBert-1.3B-NLI 自然语言理解

- Github: [Fengshenbang-LM](https://github.com/IDEA-CCNL/Fengshenbang-LM)
- Docs: [Fengshenbang-Docs](https://fengshenbang-doc.readthedocs.io/)

## 简介 Brief Introduction

2021年登顶FewCLUE和ZeroCLUE的中文BERT，在数个推理任务微调后的版本

This is the fine-tuned version of the Chinese BERT model on several NLI datasets, which topped FewCLUE and ZeroCLUE benchmark in 2021

## 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  | 自然语言理解 NLU | 二郎神 Erlangshen | MegatronBert |      1.3B      |    自然语言推断 NLI     |

## 模型信息 Model Information

基于[Erlangshen-MegatronBert-1.3B](https://huggingface.co/IDEA-CCNL/Erlangshen-MegatronBert-1.3B)，我们在收集的4个中文领域的NLI（自然语言推理）数据集，总计1014787个样本上微调了一个NLI版本。

Based on [Erlangshen-MegatronBert-1.3B](https://huggingface.co/IDEA-CCNL/Erlangshen-MegatronBert-1.3B), we fine-tuned a NLI version on 4 Chinese Natural Language Inference (NLI) datasets, with totaling 1,014,787 samples.

### 下游效果 Performance

|    模型 Model   | cmnli    |  ocnli  | snli    |
| :--------:    | :-----:  | :----:  | :-----:   | 
| Erlangshen-Roberta-110M-NLI | 80.83     |   78.56    | 88.01      |
| Erlangshen-Roberta-330M-NLI | 82.25      |   79.82    | 88.00 |
| Erlangshen-MegatronBert-1.3B-NLI | 84.52      |   84.17    | 88.67      |  


## 示例代码
```python
from modelscope.pipelines import pipeline
from modelscope.utils.constant import Tasks

pipeline_ins = pipeline(
		'fill-mask',
		model='Fengshenbang/Erlangshen-MegatronBert-1.3B-NLI',
        model_revision='v1.0.0'
)

print(pipeline_ins('生活的真谛是[MASK]。'))
```

## 引用 Citation

如果您在您的工作中使用了我们的模型，可以引用我们的[论文](https://arxiv.org/abs/2209.02970)：

If you are using the resource for your work, please cite the our [paper](https://arxiv.org/abs/2209.02970):

```text
@article{fengshenbang,
  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}
}
```

也可以引用我们的[网站](https://github.com/IDEA-CCNL/Fengshenbang-LM/):

You can also cite our [website](https://github.com/IDEA-CCNL/Fengshenbang-LM/):

```text
@misc{Fengshenbang-LM,
  title={Fengshenbang-LM},
  author={IDEA-CCNL},
  year={2021},
  howpublished={\url{https://github.com/IDEA-CCNL/Fengshenbang-LM}},
}
```