gates04/tokenized-network-intrusion-dataset

该数据集包含四个主要特征：input_ids（int32类型的序列）、token_type_ids（int8类型的序列）、attention_mask（int8类型的序列）和labels（int64类型的值）。数据集分为训练集、验证集和测试集，其中训练集包含1,187,781个样本，大小为3,672,618,852字节；验证集和测试集各包含254,525个样本，大小均为786,991,300字节。数据集的总大小为5,246,601,452字节，下载大小为570,730,378字节。

The dataset includes four main features: input_ids (sequence of int32), token_type_ids (sequence of int8), attention_mask (sequence of int8), and labels (int64 values). The dataset is split into training, validation, and test sets, with the training set containing 1,187,781 samples and sized at 3,672,618,852 bytes; both the validation and test sets contain 254,525 samples each, with a size of 786,991,300 bytes each. The total size of the dataset is 5,246,601,452 bytes, and the download size is 570,730,378 bytes.