🚩 Report: Spam
This is not a dataset.
This is an advertisement for a dataset. They refer people to their website to purchase.
they didn't post any data in here - only a few samples that they ruined with bounding boxes.
This repo is purely an advertisement, with no data useful for any purpose.
They even explicitly name the images "demo" (as opposed to actual data) as well as drawing bounding boxes on the image to make it unusable.
Huggingface is a platform for publishing and sharing, not advertising. Your practice of using it for advertising a closed dataset is not acceptable.
agree it's not clear from the readme so I've opened this pull request to hopefully make it clearer: https://huggingface.co/datasets/Nexdata/5147_Images_Japanese_Handwriting_OCR_data/discussions/3
thanks, appreciate your responsiveness here!