MahmoodAnaam/ok-vqa-ar-en-2

该数据集包含图像、问题和答案，主要用于视觉问答任务。数据集分为训练集和验证集，训练集包含9009个样本，验证集包含5046个样本。每个样本包含图像、问题和答案。问题数据包含英文和阿拉伯文两种语言，答案数据也包含英文和阿拉伯文，并且包含原始文本和置信度等信息。图像数据以image类型存储。

This dataset contains images, questions, and answers, primarily used for visual question answering tasks. The dataset is divided into a training set and a validation set, with the training set containing 9009 samples and the validation set containing 5046 samples. Each sample includes an image, a question, and answers. The question data includes both English and Arabic languages, and the answer data also includes English and Arabic, along with raw text and confidence information. The image data is stored as image type.