search

AlibabaResearch / AdvancedLiterateMachinery

A collection of original, innovative ideas and algorithms towards Advanced Literate Machinery. This project is maintained by the OCR Team in the Language Technology Lab, Tongyi Lab, Alibaba Group.
Apache License 2.0
1.15k stars 143 forks source link

关于关键信息抽取任务KIE，没有在项目中找到相关信息 #139

Open kkiskkk opened 1 month ago

kkiskkk commented 1 month ago

请问可以指出具体路径吗？ 对于example.py是否有推理脚本？ python example.py general_text_reading # task: general text reading (dump supports both image and JSON file) python example.py table_parsing # task: table parsing (dump supports both image and JSON file) python example.py formula_recognition # task: formula recognition (dump supports only JSON file) python example.py document_structurization # task: document structurization (dump supports both image and JSON file) python example.py whole_pdf_conversion # task: whole PDF conversion, i.e., converting all pages of a PDF file into an organized JSON structure (dump supports only JSON file)

kkiskkk commented 1 month ago

也没有找到pretrianed model文件夹在哪里 Training Download the swin_base_patch4_window7_224_22k.pth from Swin-Transformer and put it in pretrained_weights folder. Refer to train.sh for pretraining and finetuning.

如果但张图该如何测试？