Werea-co/werea-tr-doc-ocr-synthetic
Werea Turkish Enterprise Documents 📄🇹🇷 Türkçe kurumsal belge OCR eğitimi için tamamı sentetik sayfa görüntüleri ve birebir eşleşen markdown ground-truth metinleri. Werea tarafından Werea-DocOCR modellerinin eğitimi için üretilmiştir. Belge türleri Tür Train Test İçerik Genel vekaletname 500 50 Noter başlığı, taraflar, yetki maddeleri, noter şerhi DASK poliçesi 500 50 Poliçe/sigortalı/bina bilgileri, prim tablosu e-Arşiv fatura 500 50 Satıcı/alıcı… See the full description on the dataset page: https://huggingface.co/datasets/Werea-co/werea-tr-doc-ocr-synthetic.
Add synthetic Turkish enterprise document OCR dataset (3000 train + 300 test + generator/train scripts) (part 9)
Add synthetic Turkish enterprise document OCR dataset (3000 train + 300 test + generator/train scripts) (part 8)
Add synthetic Turkish enterprise document OCR dataset (3000 train + 300 test + generator/train scripts) (part 7)
Add synthetic Turkish enterprise document OCR dataset (3000 train + 300 test + generator/train scripts) (part 6)
Add synthetic Turkish enterprise document OCR dataset (3000 train + 300 test + generator/train scripts) (part 5)
Add synthetic Turkish enterprise document OCR dataset (3000 train + 300 test + generator/train scripts) (part 4)
Add synthetic Turkish enterprise document OCR dataset (3000 train + 300 test + generator/train scripts) (part 3)
Add synthetic Turkish enterprise document OCR dataset (3000 train + 300 test + generator/train scripts) (part 2)
Add synthetic Turkish enterprise document OCR dataset (3000 train + 300 test + generator/train scripts)
initial commit