metadata

license: mit
language:
  - fr
task_categories:
  - image-to-text
pretty_name: PELLET Casimir Marius
dataset_info:
  features:
    - name: image
      dtype: image
    - name: text
      dtype: string
  splits:
    - name: train
      num_examples: 842
      data_files: parquet/train.parquet
    - name: validation
      num_examples: 122
      data_files: parquet/validation.parquet
    - name: test
      num_examples: 125
      data_files: parquet/test.parquet
  dataset_size: 1089
tags:
  - atr
  - htr
  - ocr
  - historical
  - handwritten

PELLET Casimir Marius - Line level

PELLET Casimir Marius - Line level

Dataset Description

Homepage: Europeana
Point of Contact: TEKLIA

Dataset Summary

The PELLET Casimir Marius dataset includes 100 annotated French letters written between 1914 and 1918. Annotations were done at line-level and all images do not have any text.

Note that all images are resized to a fixed height of 128 pixels.

Languages

All the documents in the dataset are written in French.

Dataset Structure

Data Instances

{
  'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=1684x128 at 0x1A800E8E190,
  'text': 'LE HAVRE - panorama de la rue de Paris'
}

Data Fields

image: a PIL.Image.Image object containing the image. Note that when accessing the image column (using dataset[0]["image"]), the image file is automatically decoded. Decoding of a large number of image files might take a significant amount of time. Thus it is important to first query the sample index before the "image" column, i.e. dataset[0]["image"] should always be preferred over dataset["image"][0].
text: the label transcription of the image.

Usage with the PyLaia library

clone the repository or at least the data folder
you can use this dataset to reproduce the training, train a new model, infer on any of the set

Datasets:

Teklia
/

PELLET-Casimir-Marius-line