Get hundred of million of image+url from the crawling at home dataset and preprocess them - View it on GitHub