Datasets:
metadata
language:
- en
tags:
- biochemistry
- chemistry
- OCR
pretty_name: Biochemistry OCR
size_categories:
- 1K<n<10K
license: mpl-2.0
Biochemistry OCR
This is a dataset designed to fine-tune TrOCR on biochemistry terms such as Kcat/Km, various chemical names, and greek letters.
The dataset contains two columns: image
which is the path to the image, and label
which is the ground-truth text in the image.