模型:
naver-clova-ix/donut-base-finetuned-docvqa
Donut模型在DocVQA上进行了微调。它是由Geewok等人在论文 OCR-free Document Understanding Transformer 中提出的,并在 this repository 中首次发布。
声明:发布Donut模型的团队没有为该模型编写模型卡片,因此此模型卡片是由Hugging Face团队编写的。
Donut由视觉编码器(Swin Transformer)和文本解码器(BART)组成。给定一张图像,编码器首先将图像编码为嵌入张量(形状为batch_size,seq_len,hidden_size),然后解码器根据编码器的编码自动回归生成文本。
该模型在DocVQA上进行了微调,这是一个文档视觉问答数据集。
我们参考了 documentation 的代码示例。
@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}
}