lukbl/LaTeX-OCR-dataset
LaTeX-OCR Dataset Summary This dataset was created to train LaTeX-OCR, a model for recognizing LaTeX code from images of mathematical formulas. Each sample consists of a synthetically rendered formula image and its corresponding LaTeX formula. The images were generated from scratch using xelatex with multiple fonts, offering more typographic variety than many other datasets that use a single font (typically Computer Modern). Data Sources Formulas… See the full description on the dataset page: https://huggingface.co/datasets/lukbl/LaTeX-OCR-dataset.
Update README.md
Upload README.md with huggingface_hub
Upload data/validation-00000-of-00001-79453adb1121b347.parquet with huggingface_hub
Upload README.md with huggingface_hub
Upload data/train-00000-of-00001-5cb175040388de1e.parquet with huggingface_hub
initial commit
