dddraxxx/train_laion_ocr_relation_simple_5439

该数据集包含以下字段：图片（不进行解码）、数据源（字符串类型）、奖励模型（包含地面真实信息的字符串）、额外信息（包含答案、正确选项、高度、索引、检测数量、小目标检测数量、问题、样本ID、目标实例（包含边界框、实例ID、名称）、文本1、文本2、宽度）。数据集分为训练集，共有5439个示例，大小为1,261,706,315字节。额外提供了默认配置，指定了训练集的数据文件路径。

The dataset includes the following fields: images (not decoded), data source (string type), reward model (containing ground truth information as string), extra info (including answer, correct option, height, index, number of detections, number of small detections, question, sample ID, target instances (including bounding box, instance ID, name), text1, text2, width). The dataset is split into a training set with a total of 5,439 examples, sized at 1,261,706,315 bytes. A default configuration is provided, specifying the data file paths for the training set.