metadata
dataset_info:
features:
- name: query
dtype: string
- name: image
dtype: image
- name: image_filename
dtype: string
- name: text_description
dtype: string
splits:
- name: test
num_bytes: 23749713
num_examples: 102
download_size: 7706685
dataset_size: 23749713
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
Beverages Catalogue RU
This dataset was self-curated by searching beverage catalogs on Google search and downloading PDFs. Queries were synthetically generated using Qwen2b. The text_description column contains OCR text extracted from the images using EasyOCR.
Copyright Notice
We do not own the original text or image content in this dataset. If you own content in this dataset and wish to have it removed due to copyright or intellectual property concerns, please get in touch with us at "support-data (at) jina.ai" with clear identification of yourself, the copyrighted work, and the specific infringing content. We will promptly review and address legitimate requests in accordance with the applicable copyright laws.
Copyright
All rights are reserved to the original authors of the documents.