README.md

---
language:
- sv
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
dataset_info:
  features:
  - name: batch_name
    dtype: string
  - name: image_name
    dtype: string
  - name: image
    dtype: image
  - name: txt
    dtype: string
  - name: alto_raw
    dtype: string
  - name: alto_base64
    dtype: string
  splits:
  - name: train
    num_bytes: 1059494490.0
    num_examples: 824
  download_size: 1014753189
  dataset_size: 1059494490.0
---


# Dataset Card for Linköping

This is temporary datasets ..


---

## Dataset Structure

The dataset consists of five main columns:

- **`batch_name`**: The name of the batch or collection the document belongs to (e.g., `P1863`).
- **`image_name`**: The name of the specific document image (e.g., `P1863_00001`).
- **`image`**: The binary content of the document's JPEG image.
- **`txt`**: The extracted plain text content from the corresponding XML file.
- **`alto`**: The binary content of the original ALTO XML file.

### Example Entry

| batch_name | image_name    | image             | txt                                | alto               |
|------------|---------------|-------------------|------------------------------------|--------------------|
| `P1863`    | `P1863_00001` | Binary (JPEG)     | `"hvarom särskild skrifvelse..."` | Binary (XML)       |

---