Pix2struct Docvqa Large

🚀 Model card for Pix2Struct - Finetuned on Doc-VQA (Visual Question Answering over scanned documents) - large version

Pix2Struct is an image encoder - text decoder model. It's trained on image - text pairs, useful for tasks like image captioning and visual question answering. This finetuned large - version model is designed for Visual Question Answering over scanned documents.

📚 Table of Contents

0. TL;DR
1. Using the model
2. Contribution
3. Citation

📋 TL;DR

Pix2Struct is an image encoder - text decoder model trained on image - text pairs for various tasks, including image captioning and visual question answering. The full list of available models can be found in Table 1 of the paper:

The abstract of the model states that:

Visually - situated language is ubiquitous—sources range from textbooks with diagrams to web pages with images and tables, to mobile apps with buttons and forms. Perhaps due to this diversity, previous work has typically relied on domain - specific recipes with limited sharing of the underlying data, model architectures, and objectives. We present Pix2Struct, a pretrained image - to - text model for purely visual language understanding, which can be finetuned on tasks containing visually - situated language. Pix2Struct is pretrained by learning to parse masked screenshots of web pages into simplified HTML. The web, with its richness of visual elements cleanly reflected in the HTML structure, provides a large source of pretraining data well - suited to the diversity of downstream tasks. Intuitively, this objective subsumes common pretraining signals such as OCR, language modeling, image captioning. In addition to the novel pretraining strategy, we introduce a variable - resolution input representation and a more flexible integration of language and vision inputs, where language prompts such as questions are rendered directly on top of the input image. For the first time, we show that a single pretrained model can achieve state - of - the - art results in six out of nine tasks across four domains: documents, illustrations, user interfaces, and natural images.

💻 Using the model

🚚 Converting from T5x to huggingface

You can use the convert_pix2struct_checkpoint_to_pytorch.py script as follows:

python convert_pix2struct_checkpoint_to_pytorch.py --t5x_checkpoint_path PATH_TO_T5X_CHECKPOINTS --pytorch_dump_path PATH_TO_SAVE

If you are converting a large model, run:

python convert_pix2struct_checkpoint_to_pytorch.py --t5x_checkpoint_path PATH_TO_T5X_CHECKPOINTS --pytorch_dump_path PATH_TO_SAVE --use-large

Once saved, you can push your converted model with the following snippet:

from transformers import Pix2StructForConditionalGeneration, Pix2StructProcessor

model = Pix2StructForConditionalGeneration.from_pretrained(PATH_TO_SAVE)
processor = Pix2StructProcessor.from_pretrained(PATH_TO_SAVE)

model.push_to_hub("USERNAME/MODEL_NAME")
processor.push_to_hub("USERNAME/MODEL_NAME")

▶️ Running the model

The instructions for running this model are totally similar to the instructions stated on the [pix2struct - aid - base](https://huggingface.co/ybelkada/pix2struct - ai2d - base) model.

👥 Contribution

This model was originally contributed by Kenton Lee, Mandar Joshi et al. and added to the Hugging Face ecosystem by Younes Belkada.

📖 Citation

If you want to cite this work, please consider citing the original paper:

@misc{https://doi.org/10.48550/arxiv.2210.03347,
  doi = {10.48550/ARXIV.2210.03347},
  
  url = {https://arxiv.org/abs/2210.03347},
  
  author = {Lee, Kenton and Joshi, Mandar and Turc, Iulia and Hu, Hexiang and Liu, Fangyu and Eisenschlos, Julian and Khandelwal, Urvashi and Shaw, Peter and Chang, Ming - Wei and Toutanova, Kristina},
  
  keywords = {Computation and Language (cs.CL), Computer Vision and Pattern Recognition (cs.CV), FOS: Computer and information sciences, FOS: Computer and information sciences},
  
  title = {Pix2Struct: Screenshot Parsing as Pretraining for Visual Language Understanding},
  
  publisher = {arXiv},
  
  year = {2022},
  
  copyright = {Creative Commons Attribution 4.0 International}
}