An open-access, systematically curated dataset of 33 ancient Pyu handwritten consonants (825 original + 4,950 augmented samples) in CSV and Parquet formats to support ML, OCR, and digital humanities. Inspired by Thiripyanchi U Tha Myat's canonical works.
machine-learning ocr computer-vision dataset optical-character-recognition ancient-scripts pyu-script myanmar-heritage
-
Updated
Jul 17, 2026