Overview

LayoutLMv2-base-uncased-finetuned-docvqa is a multimodal transformer designed for Document Visual Question Answering (DocVQA). Unlike standard NLP models, it processes text, layout coordinates, and image embeddings simultaneously, allowing it to understand the spatial relationships within scanned documents, invoices, and forms. For developers, this means the model can pinpoint answers based on visual cues—like finding a total amount in a specific corner of a receipt—rather than relying solely on text sequences. It is an ideal choice for automating data extraction from semi-structured documents where visual positioning is critical for semantic meaning. It integrates well into OCR pipelines and is a significant step up from text-only models for any RAG system involving PDFs or images.

Highlights

Combines text, layout, and image embeddings for multimodal understanding
Specialized for high-accuracy Document Visual Question Answering tasks
Excels at extracting data from semi-structured forms and invoices
Outperforms text-only models by leveraging spatial document coordinates

Usage

Install

# Install Hugging Face transformers
pip install transformers torch

SDK Usage

# Load model with transformers
from transformers import AutoModel, AutoTokenizer

model = AutoModel.from_pretrained("tiennvcs/layoutlmv2-base-uncased-finetuned-docvqa")
tokenizer = AutoTokenizer.from_pretrained("tiennvcs/layoutlmv2-base-uncased-finetuned-docvqa")

Hugging Face Download

We recommend downloading the model via the Hugging Face CLI or Hub SDK.

Guidance：Before downloading, install huggingface_hub with:

Guidance

pip install -U huggingface_hub

CLI Download

Download the full repository

huggingface-cli download tiennvcs/layoutlmv2-base-uncased-finetuned-docvqa

Download a single file to a local folder (e.g. config.json into ./dir)

huggingface-cli download tiennvcs/layoutlmv2-base-uncased-finetuned-docvqa config.json --local-dir ./dir

See the official docs for more CLI options

SDK Download

# 模型下载
from huggingface_hub import snapshot_download
model_dir = snapshot_download('tiennvcs/layoutlmv2-base-uncased-finetuned-docvqa')

Git Download

Make sure git-lfs is installed first

Git Download

git lfs install
git clone https://huggingface.co/tiennvcs/layoutlmv2-base-uncased-finetuned-docvqa

To skip LFS large-file downloads, use:

Skip LFS

GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/tiennvcs/layoutlmv2-base-uncased-finetuned-docvqa

Model files are hosted on the Hugging Face Hub — download directly via HF CLI / SDK / Git, not through this site.

PyTorch / Transformers Usage

Install Transformers

pip install -U transformers torch

Load the model and run inference

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained('tiennvcs/layoutlmv2-base-uncased-finetuned-docvqa')
tokenizer = AutoTokenizer.from_pretrained('tiennvcs/layoutlmv2-base-uncased-finetuned-docvqa')

Full Documentation

来源: HuggingFace

---
license: cc-by-sa-4.0
tags:

generated_from_trainer

model-index:

name: layoutlmv2-base-uncased-finetuned-docvqa

results: []
---