datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
funsdDescriptionThe FUNSD (Form Understanding in Noisy Scanned Documents) dataset is a benchmark for form understanding.
Content
199 scanned forms
Entities annotated with 4 categories: header, question, answer, other
OCR text + layout information available
Forms are noisy and semi-structured
Tests ability to capture hierarchical relationships between entities
Intended Use
NER/KIE for form documents
LicenseProvided by FUNSD dataset authors (2019).
usa_passports_ocr_funsd