---
title: hwtcm-sft-v1
canonical_url: "https://www.modelscope.cn/datasets/huangxp/hwtcm-sft-v1"
md_url: "https://www.modelscope.cn/datasets/huangxp/hwtcm-sft-v1.md"
repository: huangxp/hwtcm-sft-v1
chinese_name: "传统中医SFT数据集"
last_updated: 2025-03-05
license: "Apache License 2.0"
storage_size: "3.4 MB"
downloads: 3113
stars: 17
---

# hwtcm-sft-v1

> hwtcm-sft-v1 - huangxp 在 ModelScope 开源的数据集。可用于微调传统中医领域大模型。

huangxp/hwtcm-sft-v1 是 ModelScope 魔搭社区上的数据集，存储大小 3.4 MB，采用 Apache License 2.0 许可。

- **Repository**: huangxp/hwtcm-sft-v1
- **License**: Apache License 2.0
- **Storage size**: 3.4 MB
- **Downloads**: 3113
- **Stars**: 17
- **Last updated**: 2025-03-05

Source: https://www.modelscope.cn/datasets/huangxp/hwtcm-sft-v1

---

# A dataset of Tradictional Chinese Medicine (TCM) for SFT
一个用于微调LLM的传统中医数据集

## Introduction

This repository contains a dataset of Traditional Chinese Medicine (TCM) for fine-tuning large language models.

## Dataset Description

The dataset contains 7,096 Chinese sentences related to TCM. The sentences are collected from various sources on the Internet, including medical websites, TCM forums, and TCM books. The dataset is generated or judged by various LLMs, including llama3.1-405b, glm4-flash, lmternlm2.5-latest, qwen1.5-7b, glm4-9b, glm3-6b, yi1.5-9b, yi1.5-6b and internlm2.5-7b.  
包含7096个QA，语料收集于中医网站、论坛、书籍等，然后由llama3.1-405b、glm4-flash、lmternlm2.5-latest、qwen1.5-7b、glm4-9b、glm3-6b、yi1.5-9b、yi1.5-6b、internlm2.5-7b模型转换成QA，最后同样由这些模型从专业性、安全性、流畅性三个维度判断质量以及判断分类。

## Statistics

![alt text](images/categories.png "Title")

![alt text](images/scores.png "Title")

## Limitations

The dataset is not perfect, and there may be some errors or biases. We will continue to improve the dataset and provide more accurate and balanced data.  
数据集分布不均衡，生成、质量评分以及分类完全通过LLMs完成。

## Citation

If you find this project useful in your research, please consider cite:

```
@misc{hwtcm2024,
    title={{hwtcm-sft} A traditional Chinese medicine QA dataset for fine-tuning large language models},
    author={Haiwei AI Team},
    howpublished = {\url{https://huggingface.co/datasets/Monor/hwtcm-sft-v1}},
    year={2024}
}
```