remyxai/SpaceOm_flash

这是一个包含图像和文本数据的多模态数据集。数据集中的图像被存储在images字段中，文本信息则存储在messages字段内的content子字段中，每个文本条目包括索引(index)、文本内容(text)和类型(type)。此外，数据集还包含了推理(reasoning)、输入(input)和输出(output)字段。数据集被划分为训练集(train)，包含7534个示例，总大小约为310.94MB。提供了默认配置，可通过data/train-*路径访问训练数据。

This is a multimodal dataset containing images and text. Images are stored in the images field, while text information is stored within the content subfield of the messages field, with each text entry including an index, text content, and type. Additionally, the dataset includes reasoning, input, and output fields. The dataset is split into a training set (train) with 7534 examples, totaling approximately 310.94MB in size. A default configuration is provided, with training data accessible via the data/train-* path.