mmembed/test_vqa
收藏官方服务：
资源简介：
这是一个包含文本和图像数据的训练集，共有762873个样本，数据集大小为约47GB。数据集仅包含训练集分割，可通过默认配置访问训练数据。
This dataset contains text and image data, with a total of 762,873 samples and a dataset size of approximately 47GB. The dataset includes only a training split, and the training data can be accessed through the default configuration.
提供机构：
mmembed
