Shayanvsf/Entailment

该数据集包含三个字段：前提（premise）、假设（hypothesis）和标签（label），均为文本类型。它被划分为训练集、验证集、测试集和异常分布（OOD）集，分别用于模型训练、验证、测试以及检测模型对于非训练分布数据的处理能力。数据集的具体大小和示例数量如下：训练集包含23097个示例，大小为4606527字节；验证集包含1304个示例，大小为260422字节；测试集包含2126个示例，大小为410267字节；OOD集包含5000个示例，大小为518696字节。数据集的总大小为5795912字节，下载大小为1969082字节。

The dataset includes three fields: premise, hypothesis, and label, all of which are text types. It is divided into training set, validation set, test set, and out-of-distribution (OOD) set for model training, validation, testing, and detecting the models ability to handle data outside the training distribution. The specific size and number of examples in the dataset are as follows: the training set contains 23097 examples and is 4606527 bytes in size; the validation set contains 1304 examples and is 260422 bytes in size; the test set contains 2126 examples and is 410267 bytes in size; the OOD set contains 5000 examples and is 518696 bytes in size. The total size of the dataset is 5795912 bytes, and the download size is 1969082 bytes.