Dataset Viewer
The dataset viewer is not available for this dataset.
Unexpected token '<', "<html> <h"... is not valid JSON

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

Nameplates OCR

Image-to-text dataset of nameplates, merged from local folders and/or Hugging Face datasets and built with create_dataset.py. Text-less images were auto-captioned with EasyOCR; sources with a text/metadata column use that text (flattened to plain-text lines for consistency).

Dataset structure

Split Images
train 24
test 6

Each row: {"image": <PIL.Image>, "text": "..."}.

Usage

from datasets import load_dataset

dataset = load_dataset("darjja/nameplate_sample2")
print(dataset["train"][0])
Downloads last month
38