本项目基于 200 万条医疗数据，对 DeepSeek-R1-Distill-Qwen-32B 模型进行微调，打造出一款在医学领域具备极高专业性的可本地部署大语言模型。该模型适用于医疗研究、辅助诊断等场景，支持灵活的本地运行和便捷的调用。
This project fine-tunes the DeepSeek-R1-Distill-Qwen-32B model using a dataset of 2 million medical records, delivering a highly specialized, locally deployable large language model tailored for the medical domain. It is ideal for medical research, diagnostic assistance, and more, with flexible local deployment and easy integration.
-
运行笔记本：通过 IPython Notebook (ipynb) 文件自行运行模型。
-
直接调用：访问现成版本 Hugging Face 模型仓库。
-
Run it yourself: Use the provided IPython Notebook (ipynb).
-
Ready-to-use: Access the pre-built version on Hugging Face.
| 日期 | 更新内容 |
|---|---|
| 2025.2.13 | 更新部署代码，新增简洁 UI 界面 |
| 2025.2.19 | 更新 ipynb 文件，全面优化模型参数 |
| 2025.2.25 | 升级至 DeepSeek-R1-Distill-Qwen-32B，提升专业性，减少 Bug，优化 UI |
- February 13, 2025: Updated deployment code with a simple UI interface.
- February 19, 2025: Updated ipynb with comprehensive parameter optimization.
- February 25, 2025: Upgraded to DeepSeek-R1-Distill-Qwen-32B, enhancing professionalism, reducing bugs, and improving UI.
run ipynb