doclayout-yolo-slim

DocLayout-YOLO-Slim

This library is just a lightweight slim version of the original doclayout-yolo library focued on inference of the models that were developed by OpenDataLab.

Installation

From PyPI (Coming Soon)

pip install doclayout-yolo-slim

From Source

git clone https://github.com/yourusername/doclayout-yolo-slim.git
cd doclayout-yolo-slim
pip install -e .

Using uv

uv add doclayout-yolo-slim

Quick Start

from doclayout_yolo_slim.models import YOLOv10

# Load the model
model = YOLOv10(model="doclayout_yolo_docsynth300k_imgsz1600.pt")

# Run inference
results = model.predict("path/to/your/image.png")
print(results)

Model Files

You'll need the pre-trained model file with original library. The example uses doclayout_yolo_docsynth300k_imgsz1600.pt which should be placed in your project directory or specify the full path.

Requirements

• Python >= 3.11
• PyTorch >= 2.7.1
• OpenCV >= 4.11.0
• NumPy >= 2.3.1
• Other dependencies listed in pyproject.toml

Performance

This slim implementation offers:

• Reduced memory usage
• Faster inference times
• Smaller package size
• Simplified codebase for easier maintenance

License

This project is licensed under the AGPL-3.0 License - see the LICENSE file for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Acknowledgments

• Based on the original ultralytics YOLO implementation
• Inspired by doclayout-yolo for document layout analysis
• Optimized for production use cases requiring speed and efficiency

Changelog

v0.1.0

• Initial release
• Simplified YOLOv10 implementation