Pytorch实现NSS-KDD预处理

1. 导入必要的库和模块

import pandas as pd
import numpy as np
import torch
from sklearn.preprocessing import LabelEncoder

2. 加载数据集

train = pd.read_csv('dataset/KDDTrain+.txt', header=None)
test = pd.read_csv('dataset/KDDTest+.txt', header=None)

3. 对数据进行处理

# 合并训练集和测试集，方便一起处理
data = pd.concat([train, test], axis=0)

# 删除无用的列
data.drop([41], axis=1, inplace=True)

# 对类别特征进行编码
cat_cols = [1, 2, 3]
for col in cat_cols:
    label_enc = LabelEncoder()
    data.iloc[:, col] = label_enc.fit_transform(data.iloc[:, col])

# 对标签进行编码
label_enc = LabelEncoder()
data.iloc[:, -1] = label_enc.fit_transform(data.iloc[:, -1])

# 将数据转换为numpy数组
data = data.to_numpy()

# 将数据分为特征和标签
x, y = data[:, :-1], data[:, -1]

# 将数据转换为PyTorch张量
x = torch.tensor(x, dtype=torch.float32)
y = torch.tensor(y, dtype=torch.long)

4. 划分训练集和测试集

# 划分训练集和测试集
train_size = int(0.8 * len(x))
test_size = len(x) - train_size
train_x, test_x = torch.utils.data.random_split(x, [train_size, test_size])
train_y, test_y = torch.utils.data.random_split(y, [train_size, test_size])

5. 数据标准化

# 计算训练集的均值和标准差
mean = torch.mean(train_x, dim=0)
std = torch.std(train_x, dim=0)

# 将训练集和测试集标准化
train_x = (train_x - mean) / std
test_x = (test_x - mean) / std

6. 将数据集封装成PyTorch数据集

train_dataset = torch.utils.data.TensorDataset(train_x, train_y)
test_dataset = torch.utils.data.TensorDataset(test_x, test_y)