ButterChicken98/soyabean_aerial_plus_healthy_rag_may_k2

该数据集是一个图像-文本数据集，包含1400个训练样本，每个样本包括图像、标题、多个标签以及RAG（检索增强生成）相关的字段，如检索到的块ID、方面、分数和源语料库。数据集主要用于支持图像描述、分类或检索增强生成任务，但具体应用场景未在README中明确说明。This dataset is an image-text dataset containing 1400 training samples. Each sample includes an image, captions, multiple labels, and RAG (Retrieval-Augmented Generation) related fields such as retrieved chunk IDs, aspects, scores, and source corpus. The dataset is primarily designed to support tasks like image captioning, classification, or retrieval-augmented generation, but specific application scenarios are not explicitly mentioned in the README.