wnkh/IOC download history
wnkh/IOC is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 11 times (2 in the last 7 days), and 128 times in total. It ranks #540,266 among datasets by monthly downloads.
Interpres OCR Correction (IOC) is the dataset that is desgined for training ByT5 models to correct transcribed Medieval Latin texts. The data is collected from the OCR results from multiple OCR models/engines like Deepseek-OCR, PaddleOCR-v5, TrOCR-Medieval, TRIDIS, ... on the CATMuS Latin dataset. T