xzc/pdf_title_image @ 1644b5037582f30e5c92befb255d045d29736e03

Ingen beskrivning

sprivacy 1644b50375 merge lfy branch		1 år sedan
extracted_images	76f6e51fc3 version 0.1	1 år sedan
LLMAgent.py	9365b80daa 08/08/2024 16:12:52	1 år sedan
README.md	db72ca3de4 08/16/2024 10:49:47	1 år sedan
document_.py	6ece768b5c document_.py更新	1 år sedan
extract_financial_report.py	48b9a3e9b3 merge zzh branch	1 år sedan
extract_price.py	48b9a3e9b3 merge zzh branch	1 år sedan
get_info.py	0152861ab0 08/08/2024 17:19:21	1 år sedan
instance_locate.py	48b9a3e9b3 merge zzh branch	1 år sedan
lmu.py	59d20e64a0 添加BMP图形保存，添加摘要生成、关键词检测	1 år sedan
matcher.py	65241d1460 添加评审因素段落定位	1 år sedan
ocr_api.py	48b9a3e9b3 merge zzh branch	1 år sedan
project_loc.py	db72ca3de4 08/16/2024 10:49:47	1 år sedan
requirements.txt	7e37b38fec extract content with outlines	1 år sedan
responser.py	db72ca3de4 08/16/2024 10:49:47	1 år sedan
scan_dir.py	48b9a3e9b3 merge zzh branch	1 år sedan
text_extractor.py	48b9a3e9b3 merge zzh branch	1 år sedan
tools.py	1644b50375 merge lfy branch	1 år sedan

主要模块描述 1、tools 大纲解析模块 2、get_info PDF信息抽取模块 3、matcher 段落定位模块

4、projectloc 项目业绩的表格定位模块 5、responser 格式化模块 6、lmu 摘要生成模块 7、LLMAgent 大模型调用模块 8、document 招标文件解析模块

PDF中无边框表格内容抽取

1. camelot-py git源下载
	git clone https://www.github.com/camelot-dev/camelot
	修改pyproject.toml中 pdfminer-six = "^20231228"
	安装命令： 进入camelot目录下，pip install -e .
2. 在wsl Debian中安装 ghostscript 【模块本身】
	apt install ghostscript
3. ghostscript 下载
	pip install ghostscript==0.7.0 【模块驱动】
4. 代码修改 【CV运行时不需要设置宽高，使用默认即可】
	tables_pro = camelot.read_pdf(
                    self.file_path,
                    # flavor='stream',
                    pages=str(page_number+1),
                    # edge_tol=200,
                    # row_tol=50,
                )

README.md

PDF中无边框表格内容抽取