nihalbaig/docvqa_syn_test

该数据集包含用于文档视觉问答任务的问题-答案对。数据集结构包括图像数据（存储为图像特征）、英文问题、孟加拉语答案（单个字符串）、每个问题的唯一标识符、问题类型（提取）、文档标识符和字体信息。

This dataset contains question-answer pairs for document visual question answering tasks. The dataset structure includes image data (stored as Image feature), questions in English, answers in Bengali (single string), unique identifier for each question, type of question (extraction), document identifier, and font information.