Nepali Colloquial Corpus Overview
Rows: 431,648
This subset contains conversational Nepali collected from YouTube comments.
## Repository Contents
- `README.md`: dataset card
- `LICENSE.md`: license summary
- `docs/OVERVIEW.md`: high-level summary
- `docs/DATA_SCHEMA.md`: field descriptions
- `docs/DATA_SOURCES.md`: provenance notes
- `docs/USAGE.md`: loading examples
- `docs/QUALITY_AND_LIMITATIONS.md`: dataset caveats
- `CITATION.cff`: citation metadata
- `nepali_corpus_colloquial.parquet`: primary parquet file