mugezhang/xnli-en-es-ipa
收藏官方服务:
资源简介:
这是一个文本分类数据集,包含五个字段:前提(premise)、假设(hypothesis)、标签(label)、假设的音素表示(hypothesis-phoneme)和前提的音素表示(premise-phoneme)。标签字段用于分类,包含三种类型:蕴含(entailment)、中立(neutral)和矛盾(contradiction)。数据集分为测试集、训练集和验证集,分别包含10020、785404和4980个示例。
This is a text classification dataset containing five fields: premise, hypothesis, label, hypothesis-phoneme, and premise-phoneme. The label field is used for classification and includes three types: entailment, neutral, and contradiction. The dataset is split into test, train, and validation sets, containing 10,020, 785,404, and 4,980 examples respectively.
提供机构:
mugezhang


