rinabuoy/Khmer-Flick3k-Full

该数据集包含图像及其对应的标题、句子ID、分割类型、图像ID、文件名以及高棉语标题。数据集仅包含一个测试分割，共有31,014个示例，总大小为4,354,758,255.86字节，下载大小为4,315,616,873字节。数据集的默认配置文件中指定了数据文件路径为test-*。

This dataset contains images and their corresponding caption information. The images are represented by the image field, and the caption information is represented by the caption field, which is a list of strings. Additionally, the dataset includes other metadata such as sentids (list of sentence IDs), split (dataset split type), img_id (image ID), filename (file name), and caption_khm (possibly a description in another language). The dataset is split into a test set, containing 31014 samples.