Pix2struct Ai2d Base

🚀 Pix2Struct - 在AI2D（科学图表视觉问答）上微调的模型卡

Pix2Struct是一个图像编码器 - 文本解码器模型，可处理图像文本对，适用于图像描述和视觉问答等多种任务。该模型在视觉问答任务上进行了微调，能有效处理与视觉相关的语言理解问题。

🚀 快速开始

模型简介

Pix2Struct是一个预训练的图像到文本模型，用于纯视觉语言理解，可在包含视觉语言的任务上进行微调。它通过学习将网页的掩码截图解析为简化的HTML进行预训练，网络丰富的视觉元素能很好地反映在HTML结构中，为预训练数据提供了大量来源，适合下游任务的多样性。

模型运行

在CPU上以全精度运行

你可以在CPU上以全精度运行该模型：

import requests
from PIL import Image
from transformers import Pix2StructForConditionalGeneration, Pix2StructProcessor

image_url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/ai2d-demo.jpg"
image = Image.open(requests.get(image_url, stream=True).raw)

model = Pix2StructForConditionalGeneration.from_pretrained("google/pix2struct-ai2d-base")
processor = Pix2StructProcessor.from_pretrained("google/pix2struct-ai2d-base")

question = "What does the label 15 represent? (1) lava (2) core (3) tunnel (4) ash cloud"

inputs = processor(images=image, text=question, return_tensors="pt")

predictions = model.generate(**inputs)
print(processor.decode(predictions[0], skip_special_tokens=True))
>>> ash cloud

在GPU上以全精度运行

import requests
from PIL import Image
from transformers import Pix2StructForConditionalGeneration, Pix2StructProcessor

image_url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/ai2d-demo.jpg"
image = Image.open(requests.get(image_url, stream=True).raw)

model = Pix2StructForConditionalGeneration.from_pretrained("google/pix2struct-ai2d-base").to("cuda")
processor = Pix2StructProcessor.from_pretrained("google/pix2struct-ai2d-base")

question = "What does the label 15 represent? (1) lava (2) core (3) tunnel (4) ash cloud"

inputs = processor(images=image, text=question, return_tensors="pt").to("cuda")

predictions = model.generate(**inputs)
print(processor.decode(predictions[0], skip_special_tokens=True))
>>> ash cloud

在GPU上以半精度运行

import requests
from PIL import Image

import torch
from transformers import Pix2StructForConditionalGeneration, Pix2StructProcessor

image_url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/ai2d-demo.jpg"
image = Image.open(requests.get(image_url, stream=True).raw)

model = Pix2StructForConditionalGeneration.from_pretrained("google/pix2struct-ai2d-base", torch_dtype=torch.bfloat16).to("cuda")
processor = Pix2StructProcessor.from_pretrained("google/pix2struct-ai2d-base")

question = "What does the label 15 represent? (1) lava (2) core (3) tunnel (4) ash cloud"

inputs = processor(images=image, text=question, return_tensors="pt").to("cuda", torch.bfloat16)

predictions = model.generate(**inputs)
print(processor.decode(predictions[0], skip_special_tokens=True))
>>> ash cloud

从T5x转换到Hugging Face

你可以使用convert_pix2struct_checkpoint_to_pytorch.py脚本进行转换：

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

如果你要转换大型模型，请运行：

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

保存后，你可以使用以下代码将转换后的模型推送到Hugging Face Hub：

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")

✨ 主要特性

• 多语言支持：支持英语、法语、罗马尼亚语、德语等多种语言。
• 视觉问答能力：经过微调，可处理视觉问答任务，能准确回答与图像相关的问题。
• 预训练策略新颖：通过将网页掩码截图解析为简化HTML进行预训练，适用于多种下游任务。

📦 安装指南

文档未提及具体安装步骤，可参考Hugging Face相关文档进行安装。

💻 使用示例

基础用法

在CPU上以全精度运行模型的示例：

import requests
from PIL import Image
from transformers import Pix2StructForConditionalGeneration, Pix2StructProcessor

image_url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/ai2d-demo.jpg"
image = Image.open(requests.get(image_url, stream=True).raw)

model = Pix2StructForConditionalGeneration.from_pretrained("google/pix2struct-ai2d-base")
processor = Pix2StructProcessor.from_pretrained("google/pix2struct-ai2d-base")

question = "What does the label 15 represent? (1) lava (2) core (3) tunnel (4) ash cloud"

inputs = processor(images=image, text=question, return_tensors="pt")

predictions = model.generate(**inputs)
print(processor.decode(predictions[0], skip_special_tokens=True))
>>> ash cloud

高级用法

在GPU上以半精度运行模型的示例：

import requests
from PIL import Image

import torch
from transformers import Pix2StructForConditionalGeneration, Pix2StructProcessor

image_url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/ai2d-demo.jpg"
image = Image.open(requests.get(image_url, stream=True).raw)

model = Pix2StructForConditionalGeneration.from_pretrained("google/pix2struct-ai2d-base", torch_dtype=torch.bfloat16).to("cuda")
processor = Pix2StructProcessor.from_pretrained("google/pix2struct-ai2d-base")

question = "What does the label 15 represent? (1) lava (2) core (3) tunnel (4) ash cloud"

inputs = processor(images=image, text=question, return_tensors="pt").to("cuda", torch.bfloat16)

predictions = model.generate(**inputs)
print(processor.decode(predictions[0], skip_special_tokens=True))
>>> ash cloud

📚 详细文档

模型摘要中提到：

视觉情境语言无处不在，其来源范围从带有图表的教科书到带有图像和表格的网页，再到带有按钮和表单的移动应用程序。由于这种多样性，以前的工作通常依赖于特定领域的方法，底层数据、模型架构和目标的共享有限。我们提出了Pix2Struct，这是一个预训练的图像到文本模型，用于纯视觉语言理解，可在包含视觉情境语言的任务上进行微调。Pix2Struct通过学习将网页的掩码截图解析为简化的HTML进行预训练。网络丰富的视觉元素能很好地反映在HTML结构中，为预训练数据提供了大量来源，适合下游任务的多样性。直观地说，这个目标包含了常见的预训练信号，如OCR、语言建模、图像描述。除了新颖的预训练策略，我们还引入了可变分辨率输入表示和更灵活的语言与视觉输入集成，其中问题等语言提示直接渲染在输入图像之上。我们首次表明，一个单一的预训练模型可以在四个领域（文档、插图、用户界面和自然图像）的九个任务中的六个任务中取得最先进的结果。

🔧 技术细节

文档未提及具体技术细节。

📄 许可证

本模型使用Apache 2.0许可证。

👥 贡献者

该模型最初由Kenton Lee、Mandar Joshi等人贡献，并由Younes Belkada添加到Hugging Face生态系统中。

📖 引用

如果你想引用这项工作，请引用原始论文：

@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}
}