lingvenvist/animacy-eu-original-rebalanced-fixed

该数据集包含句子、标记、动画标签、目标索引和来源五个主要字段。动画标签字段是一个序列，包含三个类标签：N、A和H。数据集分为训练集、测试集和验证集三个部分，每个部分都有相应的文件大小和示例数量。The dataset contains five features: sentences, tokens, anim_tags, target-indexes, and source. The anim_tags feature includes three categories: N, A, and H. The dataset is divided into train, test, and validation sets, containing 20181, 2999, and 368 samples respectively. The total download size of the dataset is 3099651 bytes, and the total dataset size is 6896056 bytes.