赵小蒙 4b8dbd7cfb ocr_pdf_intermediate_dict_to_markdown_with_para支持mm和nlp双模式 1 gadu atpakaļ
..
dict2md d6a5724b26 table_latex支持 1 gadu atpakaļ
filter d5dbed7325 目录重构 1 gadu atpakaļ
layout d5dbed7325 目录重构 1 gadu atpakaļ
libs f36c26565e 使用面积占比方式判断一行文本是不是在一个layoutbox里 1 gadu atpakaļ
para d3c9cb84f8 分段部分log限定在debug模式下才能输出 1 gadu atpakaļ
post_proc 71a042d9fc footnote检测逻辑更新 1 gadu atpakaļ
pre_proc fc10772503 ocr_construct_page_component 位置移动 1 gadu atpakaļ
spark 50a543ce0e s3配置信息路径更换 1 gadu atpakaļ
train_utils efed5faa53 feat: modify foot note bbox tmp 1 gadu atpakaļ
__init__.py d5dbed7325 目录重构 1 gadu atpakaļ
pdf_parse_by_model.py 7f0c734ff6 pipeline重构 1 gadu atpakaļ
pdf_parse_by_ocr.py 872cd73f4a pipeline重构 1 gadu atpakaļ
pdf_parse_for_train.py fc10772503 ocr_construct_page_component 位置移动 1 gadu atpakaļ
pipeline.py 575ca00e01 app.common依赖删除,pipeline_ocr重构 1 gadu atpakaļ
pipeline_ocr.py 4b8dbd7cfb ocr_pdf_intermediate_dict_to_markdown_with_para支持mm和nlp双模式 1 gadu atpakaļ
pipeline_txt.py 433684c646 实现多模态markdown拼装 1 gadu atpakaļ