ahmedheakl/ar_laion1_instruct

ar_laion1_instruct数据集包含多个特征字段，如样本ID、URL、文本内容、图片尺寸、相似度、哈希值等。数据集分为训练集，共有96461个样本，总大小约为5.61GB。数据集还包含了图片的美学评分、语言类型、路径和图片对象等信息。此外，数据集还包含了对话信息，包括内容和角色。该数据集可能用于图像处理、文本分析以及对话系统等领域。

The ar_laion1_instruct dataset includes various feature fields such as sample ID, URL, text content, image dimensions, similarity, hash value, etc. The dataset is split into a training set with a total of 96,461 samples, approximately 5.61GB in size. The dataset also contains information such as the aesthetic score of images, language type, path, and image objects. In addition, the dataset includes conversation information, including content and role. This dataset may be used for image processing, text analysis, and dialogue systems, among other fields.