beyoru/Viet-OCR-VQA_4

该数据集包含多个特征，包括id（整数类型）、image（图像类型）、description（字符串类型）和conversations（列表类型，包含role和content两个字符串类型的字段）。数据集的分割部分名为quarter_4，包含34,276个示例，总大小为6,472,267,634.348字节。下载大小为6,324,264,951字节。数据集的配置部分指定了默认配置，数据文件路径为data/quarter_4-*。

The dataset contains multiple features, including id (integer type), image (image type), description (string type), and conversations (list type, containing two string fields: role and content). The datasets split section is named quarter_4, containing 34,276 examples with a total size of 6,472,267,634.348 bytes. The download size is 6,324,264,951 bytes. The datasets configuration section specifies the default configuration, with data file paths as data/quarter_4-*.