ArielPorath/vcrop-smoke-test

该数据集包含三个配置：folds、ntv3_8m_smoke和raw，用于序列分类任务。folds配置提供四折交叉验证分割，每个折叠包含训练、评估和测试集，每个示例有序列字符串、整数标签和来源字符串特征。ntv3_8m_smoke配置在folds基础上增加了NLP预处理特征，包括输入ID列表、注意力掩码列表和嵌入列表，适用于基于Transformer模型的实验。raw配置包含原始训练数据，共100个示例。数据集总下载大小约为940KB，总数据集大小约为780KB。

This dataset includes three configurations: folds, ntv3_8m_smoke, and raw, designed for sequence classification tasks. The folds configuration provides four-fold cross-validation splits, each with training, evaluation, and test sets, featuring sequence strings, integer labels, and origin strings. The ntv3_8m_smoke configuration adds NLP preprocessed features such as input IDs lists, attention mask lists, and embedding lists, suitable for Transformer-based model experiments. The raw configuration contains raw training data with 100 examples. The total download size is approximately 940KB, and the total dataset size is about 780KB.