Lunzima/ocr_dataset download history
Lunzima/ocr_dataset is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 48 times (12 in the last 7 days), and 757 times in total. It ranks #273,827 among datasets by monthly downloads.
The dataset consists of synthetic OCR training content featuring a font distribution ratio of approximately 60% serif, 32% sans-serif, and 8% monospace typefaces. The text fragments are randomly selected, and the entire paragraph has no actual meaning to prevent overfitting during training. Addition
Open Lunzima/ocr_dataset on Hugging Face
Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.