iagoalves/iagoras_dataset

该数据集包含文本(text)、标签(label)、概率(probs)、元数据(metadata)、输入ID(input_ids)、token类型ID(token_type_ids)、注意力掩码(attention_mask)、长度(length)以及是否为教育领域(is_edu)等字段。数据集分为训练集(train)，共有约6053万条示例，数据集总大小约为36.37GB。提供了默认配置，训练数据文件以data/train-*的形式存在。

The dataset includes fields such as text, label, probability (probs), metadata, input IDs (input_ids), token type IDs (token_type_ids), attention mask (attention_mask), length, and whether it is in the educational field (is_edu). The dataset is split into a training set (train) with a total of approximately 6.053 million examples, and the total size of the dataset is about 36.37GB. A default configuration is provided, and the training data files exist in the form of data/train-*.