VarunGumma/VQA-OCR-Book

这是一个包含图片和对应问题的数据集，每个样本包括一个图片ID、图片、掩码ID、掩码、一个问题和一个答案。数据集分为训练集、验证集和测试集，可用于训练和评估模型在图像理解和问题回答方面的性能。

This dataset consists of images and corresponding questions, with each sample including an image ID, an image, a mask ID, a mask, a question, and an answer. The dataset is split into training, validation, and test sets, which can be used to train and evaluate models on image understanding and question answering tasks.