Datasets:

---

mohamed-khalil

/

ATHAR

like 11

This PR introduces several improvements to the ATHAR dataset card:

• Content Consolidation: Removed a large, commented-out, and redundant block of Markdown content at the beginning of the README, ensuring a single, clean, and active dataset description.
• Metadata Tagging: Added classical-arabic to the tags in the YAML metadata to enhance the dataset's discoverability for users searching for specific language variations.
• License Alignment: Corrected the license statement in the Markdown content from CC BY NC 4.0 to CC BY SA 4.0 to match the license: cc-by-sa-4.0 specified in the YAML metadata, ensuring consistency across the card. The full license link text was also updated.
• Language Code Correction: Fixed the swapped ISO 639-3 language codes for Arabic (eng to ara) and English (ara to eng) in the "Language codes" section for factual accuracy.
• Image Visibility: Uncommented an important illustrative image in the "Description" section and refined its surrounding HTML for proper rendering and improved visual appeal.
• GitHub Link Removal: Removed a commented-out GitHub link, as no active GitHub URL was provided as context for this dataset.
• Paper Link Preservation: Kept the existing Arxiv paper link, following the guideline not to replace it with a Hugging Face paper link if an Arxiv link is already present.

These updates aim to provide a more accurate, complete, and user-friendly dataset card.