metadata

language:
  - de
license: afl-3.0
task_categories:
  - image-to-text
dataset_info:
  features:
    - name: image
      dtype: image
    - name: text
      dtype: string
  splits:
    - name: train
      num_bytes: 3795163553.91
      num_examples: 10854
  download_size: 1796789422
  dataset_size: 3795163553.91
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

German handwriting

This dataset contains German handwriting images and corresponding text labels. In total, the dataset contains around 10,000 entries with handwriting from 15 different people. The data was created with the help of transcripts from school and university. The dataset was created as part of a handwriting recognition project at the FH-SWF.

How to use:

from datasets import load_dataset

dataset = load_dataset('fhswf/german_handwriting')