模型:
naver-clova-ix/donut-base-finetuned-cord-v1
Donut模型基于CORD协议进行了微调。它是由Geewok等人在 OCR-free Document Understanding Transformer 论文中介绍并于 this repository 首次发布的。
免责声明:发布Donut模型的团队未为此模型编写模型卡片,因此该模型卡片由Hugging Face团队编写。
Donut由视觉编码器(Swin Transformer)和文本解码器(BART)组成。给定一张图片,编码器首先将图片编码为嵌入张量(形状为batch_size,seq_len,hidden_size),然后解码器在编码器的编码的条件下自回归地生成文本。
该模型是在CORD(一个文档解析数据集)上进行微调的。
我们参考了包含代码示例的 documentation 。
@article{DBLP:journals/corr/abs-2111-15664, author = {Geewook Kim and Teakgyu Hong and Moonbin Yim and Jinyoung Park and Jinyeong Yim and Wonseok Hwang and Sangdoo Yun and Dongyoon Han and Seunghyun Park}, title = {Donut: Document Understanding Transformer without {OCR}}, journal = {CoRR}, volume = {abs/2111.15664}, year = {2021}, url = {https://arxiv.org/abs/2111.15664}, eprinttype = {arXiv}, eprint = {2111.15664}, timestamp = {Thu, 02 Dec 2021 10:50:44 +0100}, biburl = {https://dblp.org/rec/journals/corr/abs-2111-15664.bib}, bibsource = {dblp computer science bibliography, https://dblp.org} }