环境部署

下载

git clone https://gitclone.com/github.com/DS4SD/docling.git
conda create -n docling python=3.11
conda activate docling
pip install docling

安装模型

git clone https://www.modelscope.cn/AI-ModelScope/docling-models.git
git clone https://gitclone.com/github.com/JaidedAI/EasyOCR.gi
t

部署问题

缺少.pth文件，是去modelscope官网下载

用法

转换单个文档

将

from docling.document_converter import DocumentConvertersource = "demo1.pdf"  # PDF path or URL
converter = DocumentConverter()
result = converter.convert(source)
print(result.document.export_to_markdown())  # output: "### Docling Technical Report[...]"

解析效果

速度0.96秒/页

但是公式解析效果差