ButterChicken98/CottonWeedID15_RAG_Captions

该数据集是一个图像-文本对数据集，包含5187个训练示例，总大小约279.86 MB。特征包括图像（image）、标签（label）、随机提示（prompt_random）及其ID、检索提示（prompt_retrieval_k1）及其ID和分数、检索提示方面（prompt_retrieval_k1_aspect）、RAG提示（prompt_rag_k2_slm）及其ID和分数、RAG提示方面（prompt_rag_k2_slm_aspect）以及回退标志（prompt_rag_k2_slm_fallback）。数据集主要用于图像生成、文本到图像任务或相关NLP研究，但具体描述未在README中提及。

This dataset is an image-text pair dataset containing 5,187 training examples with a total size of approximately 279.86 MB. Features include image, label, random prompts and their IDs, retrieval prompts (prompt_retrieval_k1) with IDs and scores, retrieval prompt aspect (prompt_retrieval_k1_aspect), RAG prompts (prompt_rag_k2_slm) with IDs and scores, RAG prompt aspect (prompt_rag_k2_slm_aspect), and a fallback flag (prompt_rag_k2_slm_fallback). The dataset is primarily intended for image generation, text-to-image tasks, or related NLP research, but a detailed description is not provided in the README.