Kowsher/multitask_vqa_benchmark
收藏官方服务:
资源简介:
该数据集是一个视觉问答和图像分类的数据集,包含了用于训练和测试的多个图像以及与之相关的文本信息,如问题、答案和提示。具体应用场景可能包括图表理解、OKVQA、科学问答等。
This dataset is a visual question answering and image classification dataset, containing multiple images for training and testing, along with associated text information such as questions, answers, and prompts. The specific application scenarios may include chart understanding, OKVQA, science QA, etc.
提供机构:
Kowsher


