---
license: apache-2.0
---
<h1><a href="https://github.com/LingyvKong/OneChart">OneChart: Purify the Chart Structural Extraction via One Auxiliary Token</a></h1>

Jinyue Chen*, Lingyu Kong*, [Haoran Wei](https://scholar.google.com/citations?user=J4naK0MAAAAJ&hl=en), Chenglong Liu, [Zheng Ge](https://joker316701882.github.io/), Liang Zhao, [Jianjian Sun](https://scholar.google.com/citations?user=MVZrGkYAAAAJ&hl=en), Chunrui Han, [Xiangyu Zhang](https://scholar.google.com/citations?user=yuB-cfoAAAAJ&hl=en)

[Github](https://github.com/LingyvKong/OneChart) 
[arxiv](https://arxiv.org/abs/2404.09987)

## Quickly try the demo using hugginface：
```python
from transformers import AutoModel, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained('kppkkp/OneChart', trust_remote_code=True, use_fast=False, padding_side="right")
model = AutoModel.from_pretrained('kppkkp/OneChart', trust_remote_code=True, low_cpu_mem_usage=True, device_map='cuda')
model = model.eval().cuda()

# input your test image
image_file = 'image.png'
res = model.chat(tokenizer, image_file, reliable_check=True)
print(res)
```