README.md

---
pipeline_tag: image-text-to-text
language:
- multilingual
tags:
- got
- vision-language
- ocr2.0
- custom_code
license: apache-2.0
---

## Same model as [stepfun-ai/GOT-OCR2_0](https://huggingface.co/stepfun-ai/GOT-OCR2_0) but using custom source code:

1. Remove the `verovio` dependency since most people don't need to OCR musical annotation.
2. Allow a user to use `float16` if their GPU doesn't support `bfloat16`.
3. Updated to support `Transformers==4.48.3` so it no longer gives a bunch of warning messages.


<details><summary>ORIGINAL MODEL CARD HERE</summary>


<h1>General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Model
</h1>

[🔋Online Demo](https://huggingface.co/spaces/ucaslcl/GOT_online) | [🌟GitHub](https://github.com/Ucas-HaoranWei/GOT-OCR2.0/) | [📜Paper](https://arxiv.org/abs/2409.01704)</a> 


[Haoran Wei*](https://scholar.google.com/citations?user=J4naK0MAAAAJ&hl=en), Chenglong Liu*, Jinyue Chen, Jia Wang, Lingyu Kong, Yanming Xu,  [Zheng Ge](https://joker316701882.github.io/), Liang Zhao, [Jianjian Sun](https://scholar.google.com/citations?user=MVZrGkYAAAAJ&hl=en), [Yuang Peng](https://scholar.google.com.hk/citations?user=J0ko04IAAAAJ&hl=zh-CN&oi=ao), Chunrui Han, [Xiangyu Zhang](https://scholar.google.com/citations?user=yuB-cfoAAAAJ&hl=en)



![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/6653eee7a2d7a882a805ab95/QCEFY-M_YG3Bp5fn1GQ8X.jpeg)



## Usage
Inference using Huggingface transformers on NVIDIA GPUs. Requirements tested on python 3.10：
```
torch==2.0.1
torchvision==0.15.2
transformers==4.37.2
tiktoken==0.6.0
verovio==4.3.1
accelerate==0.28.0
```


```python
from transformers import AutoModel, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained('ucaslcl/GOT-OCR2_0', trust_remote_code=True)
model = AutoModel.from_pretrained('ucaslcl/GOT-OCR2_0', trust_remote_code=True, low_cpu_mem_usage=True, device_map='cuda', use_safetensors=True, pad_token_id=tokenizer.eos_token_id)
model = model.eval().cuda()


# input your test image
image_file = 'xxx.jpg'

# plain texts OCR
res = model.chat(tokenizer, image_file, ocr_type='ocr')

# format texts OCR:
# res = model.chat(tokenizer, image_file, ocr_type='format')

# fine-grained OCR:
# res = model.chat(tokenizer, image_file, ocr_type='ocr', ocr_box='')
# res = model.chat(tokenizer, image_file, ocr_type='format', ocr_box='')
# res = model.chat(tokenizer, image_file, ocr_type='ocr', ocr_color='')
# res = model.chat(tokenizer, image_file, ocr_type='format', ocr_color='')

# multi-crop OCR:
# res = model.chat_crop(tokenizer, image_file, ocr_type='ocr')
# res = model.chat_crop(tokenizer, image_file, ocr_type='format')

# render the formatted OCR results:
# res = model.chat(tokenizer, image_file, ocr_type='format', render=True, save_render_file = './demo.html')

print(res)


```
More details about 'ocr_type', 'ocr_box', 'ocr_color', and 'render' can be found at our GitHub.
Our training codes are available at our [GitHub](https://github.com/Ucas-HaoranWei/GOT-OCR2.0/).



## More Multimodal Projects

👏 Welcome to explore more multimodal projects of our team:

[Vary](https://github.com/Ucas-HaoranWei/Vary) | [Fox](https://github.com/ucaslcl/Fox) | [OneChart](https://github.com/LingyvKong/OneChart)

## Citation

If you find our work helpful, please consider citing our papers 📝 and liking this project ❤️！

```bib
@article{wei2024general,
  title={General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Model},
  author={Wei, Haoran and Liu, Chenglong and Chen, Jinyue and Wang, Jia and Kong, Lingyu and Xu, Yanming and Ge, Zheng and Zhao, Liang and Sun, Jianjian and Peng, Yuang and others},
  journal={arXiv preprint arXiv:2409.01704},
  year={2024}
}
@article{liu2024focus,
  title={Focus Anywhere for Fine-grained Multi-page Document Understanding},
  author={Liu, Chenglong and Wei, Haoran and Chen, Jinyue and Kong, Lingyu and Ge, Zheng and Zhu, Zining and Zhao, Liang and Sun, Jianjian and Han, Chunrui and Zhang, Xiangyu},
  journal={arXiv preprint arXiv:2405.14295},
  year={2024}
}
@article{wei2023vary,
  title={Vary: Scaling up the Vision Vocabulary for Large Vision-Language Models},
  author={Wei, Haoran and Kong, Lingyu and Chen, Jinyue and Zhao, Liang and Ge, Zheng and Yang, Jinrong and Sun, Jianjian and Han, Chunrui and Zhang, Xiangyu},
  journal={arXiv preprint arXiv:2312.06109},
  year={2023}
}
```

</details>
<br>

# Example Usage

```python
import fitz
from PIL import Image
from transformers import AutoModel, AutoTokenizer
import torch

# The following three lines are optional - removes the last remaining logging message from Transformers.
# import warnings
# from transformers import logging as transformers_logging
# transformers_logging.set_verbosity_error()

MODEL_PATH = "ctranslate2-4you/GOT-OCR2_0-Customized"  # Replace with local path if desired
tokenizer = AutoTokenizer.from_pretrained(MODEL_PATH, trust_remote_code=True)
model = AutoModel.from_pretrained(
    MODEL_PATH,
    trust_remote_code=True,
    low_cpu_mem_usage=True,
    device_map='cuda',
    use_safetensors=True,
    pad_token_id=tokenizer.convert_tokens_to_ids("<|endoftext|>")
)
model = model.eval().cuda()

def clean_repetitive_lines(text):
    """
    Removes repetitive lines from the OCR output before saving the .txt file. This is necessary because
    the model sometimes produces OCR artifacts.  All duplicates above 2 instances are removed.
    """
    lines = text.split('\n')
    cleaned_lines = []
    i = 0
    while i < len(lines):
        cleaned_lines.append(lines[i])
        repeat_count = 1
        j = i + 1
        while j < len(lines) and lines[j] == lines[i]:
            repeat_count += 1
            j += 1
        if repeat_count > 2:
            if i + 1 < len(lines):
                cleaned_lines.append(lines[i + 1])
            i = j
        else:
            i += 1
    return '\n'.join(cleaned_lines)

@torch.inference_mode()
def process_pdf_for_ocr(tokenizer, model, pdf_path):
    pdf_document = fitz.open(pdf_path)
    full_text = []
    
    for page_num in range(len(pdf_document)):
        page = pdf_document[page_num]
        zoom = 2
        matrix = fitz.Matrix(zoom, zoom)
        pix = page.get_pixmap(matrix=matrix)
        img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
        # gradio_input=True is used because we're creating images for each page of a .pdf using PyMuPDF and Pillow instead of relying on the model's internal code
        res = model.chat_crop(tokenizer, img, ocr_type='ocr', gradio_input=True)
        
        if res.strip():
            full_text.append(res)

    complete_text = '\n'.join(full_text)
    cleaned_text = clean_repetitive_lines(complete_text)
    
    with open("extracted_text_got_ocr.txt", "w", encoding="utf-8") as f:
        f.write(cleaned_text)
    
    pdf_document.close()
    print("Results have been saved to extracted_text_got_ocr.txt")

# Example usage
pdf_path = "path/to/your/pdf"
process_pdf_for_ocr(tokenizer, model, pdf_path)
```