tsystems/sharegpt4v_vqa_200k_batch5

这是一个基于图像和文本的多模态数据集，包含图像、图像路径和查询文本等特征。数据集分为训练集，共有200,000个样本。整个数据集的大小为10,135,020,909字节。数据集适用于图像到文本的任务，语言为英语，大小分类在10万到100万之间。数据集遵循CC BY NC 4.0许可，仅限非商业用途和研究目的使用。This is a multi-modal dataset based on images and text, containing features such as images, image paths, and query text. The dataset is split into a training set with a total of 200,000 samples. The entire dataset size is 10,135,020,909 bytes. The dataset is suitable for image-to-text tasks, with English as the language and a size classification between 100K and 1M. The dataset is licensed under CC BY NC 4.0, allowing only for non-commercial use and research purposes.