You need to agree to share your contact information to access this dataset

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this dataset content.

Scyrit Art

Scyrit Art is a dataset curated to enable text-to-image generation tasks. It provides a structured collection of text descriptions (title) and corresponding image URLs (image_url), which can be leveraged for training and evaluating machine learning models focused on visual content generation.

Dataset Summary

This dataset includes:

  • title: A concise textual description or headline summarizing the image.
  • image_url: The direct URL to the corresponding image.

The data is sourced from prominent news platforms to ensure a diverse range of topics, including technology, politics, sports, health, and more.

Dataset Structure

The dataset is stored in CSV format with the following columns:

  • title: String, a brief description of the image content.
  • image_url: String, a link to the image associated with the title.

Example:

title,image_url
"Tech innovations of 2023","https://example.com/images/tech2023.jpg"
"Breaking news: Global summit begins","https://example.com/images/summit.jpg"
Downloads last month
31