SEACrowd/sea-vl-crawling-metadata-non-dup

该数据集包含多个字段，如id、lang、path等，均为字符串类型；pred_en字段用于预测英文，width和height字段表示图片的宽和高，reward字段为浮点类型，表示某种奖励值。数据集分为训练集，共有404,524个示例，总文件大小为93,391,235字节。提供了默认配置，包含训练集数据文件路径。

The dataset includes fields such as id, lang, path as strings; pred_en field for English prediction, width and height fields for image dimensions, and the reward field as a float type representing some sort of reward value. The dataset is split into a training set with a total of 404,524 examples and a total file size of 93,391,235 bytes. A default configuration is provided, including the path to the data files for the training set.