Medieval Data 🏰

Welcome to the Medieval Data organization, a dedicated platform for offering datasets specifically curated for training machine learning models on medieval-specific tasks.

These datasets and models are maintained by William J.B. Mattingly

Datasets 📚

Here's a quick overview of our available datasets:

1. MGH Critical Edition Dataset: 100 annotated pages of an MGH critical edition to parse out the main body text and titles from marginalia and footers.

Models 🛡️

1. MGH Object Detection YOLOv8: Annotate an MGH critical edition to extract the main body text and titles automatically. This helps in downstream OCR with Tesseract.

Replace dataset_name with the specific name of the dataset you're interested in.

Contribute 🤝

We welcome contributions! If you have a medieval-specific dataset or have annotations that can be added, please reach out.

License 📜

All datasets in this organization are released under the CC BY 4.0 License unless specified otherwise. Please ensure to cite the original sources and the Medieval Data organization when using the datasets.