dh-unibe/image-text_hgb-kf_mixture

Dataset Card for transkribus-exports-5657-raw-xml

This dataset was created using pagexml-hf converter from Transkribus PageXML data.

Dataset Summary

This dataset contains 154 samples across 1 split(s). Mixed Historisches Grundbuch Basel and Königsfelden.

Partly based on the Königsfelden Data Set (https://zenodo.org/records/5179361). PageXML representation of the text on the charters. For the transcription guidelines, see: (https://koenigsfelden.sources-online.org/intro.html).

Geographical scope: Switzerland<br>Period: 1300-1350<br>Languages: Middle High German<br>Type of document: Documents and files<br>Provenance: State Archives of Aargau<br>

Projects Included

• —HGBFTM4_50
• —HGBFTM4_75
• —HGBFTM4_95
• —u-17_0059
• —u-17_0060
• —u-17006101
• —u-17006102
• —u-17_0065
• —u-17_0075
• —u-17_0083
• —u-17_0103
• —u-17_0104
• —u-17_0126
• —u-17_0149
• —u-17_0151
• —u-17_0152
• —u-17_0159
• —u-17_0179
• —u-17_0185a
• —u-17_0187a

Dataset Structure

Data Splits

• —train: 154 samples

Dataset Size

• —Approximate total size: 4354.51 MB
• —Total samples: 154

Features

• —image: Image(mode=None, decode=False)
• —xml_content: Value('string')
• —filename: Value('string')
• —project_name: Value('string')

Data Organization

Data is organized as parquet shards by split and project:

data/
├── <split>/
│   └── <project_name>/
│       └── <timestamp>-<shard>.parquet

The HuggingFace Hub automatically merges all parquet files when loading the dataset.

Usage

from datasets import load_dataset

# Load entire dataset
dataset = load_dataset("dh-unibe/transkribus-exports-5657-raw-xml") 

# Load specific split
train_dataset = load_dataset("dh-unibe/transkribus-exports-5657-raw-xml", split="train")