baconnier/deepsynth-qa-natural-questions

这个数据集包含了答案、图片、指令、长答案等信息的记录。每条记录还包含一些元数据，如答案类型、配置信息、估计的高度、提取的标记数量、提取方法、生成分辨率、是否有长答案和短答案等。数据集分为训练集，其中包含约12800个示例，文件大小为3260253424字节。此外，提供了默认配置，指定了训练数据的文件路径。

The dataset includes records with information such as answers, images, instructions, and long answers. Each record also contains metadata like answer type, configuration, estimated height, extracted token count, extraction method, generation resolution, whether it has long or short answers, etc. The dataset is split into a training set, which contains approximately 12800 examples and is 3260253424 bytes in size. Additionally, a default configuration is provided, specifying the file paths for the training data.