kaistlayner/tokenized_empathy_dataset

该数据集包含多个对话相关的特征字段，如对话ID、话语索引、上下文、提示、说话者索引、话语、自我评估、标签、情感标签、相关性分数、输入ID和注意力掩码。数据集分为训练集、验证集和测试集，分别包含76673、12030和10943个样本。总下载大小为13436818字节，数据集总大小为280739353字节。

The dataset includes multiple features related to dialogues, such as conversation ID, utterance index, context, prompt, speaker index, utterance, self-evaluation, tags, emotion labels, relevance scores, input IDs, and attention masks. The dataset is divided into training, validation, and test sets, containing 76673, 12030, and 10943 samples respectively. The total download size is 13436818 bytes, and the total dataset size is 280739353 bytes.