python2 计算出来的这个图片输出，咋知道有啥意义？ 怎么知道哪个是网络入侵攻击呀？

Question

# coding=utf-8 from __future__ import division import numpy as np import matplotlib.pyplot as plt def classify(input_vct, data_set): data_set_size = data_set.shape[0] diff_mat = np.tile(input_vct, (data_set_size, 1)) - data_set # 扩充input_vct到与data_set同型并相减 sq_diff_mat = diff_mat**2 # 矩阵中每个元素都平方 distance = sq_diff_mat.sum(axis=1)**0.5 # 每行相加求和并开平方根 return distance.min(axis=0) # 返回最小距离 def file2mat(test_filename, para_num): """ 将表格存入矩阵，test_filename为表格路径，para_num为存入矩阵的列数 返回目标矩阵，和矩阵每一行数据的类别 """ fr = open(test_filename) lines = fr.readlines() line_nums = len(lines) result_mat = np.zeros((line_nums, para_num)) # 创建line_nums行，para_num列的矩阵 class_label = [] for i in range(line_nums): line = lines[i].strip() item_mat = line.split(',') result_mat[i, :] = item_mat[0: para_num] class_label.append(item_mat[-1]) # 表格中最后一列正常1异常2的分类存入class_label fr.close() return result_mat, class_label def roc(data_set): normal = 0 data_set_size = data_set.shape[1] roc_rate = np.zeros((2, data_set_size)) for i in range(data_set_size): if data_set[2][i] == 1: normal += 1 abnormal = data_set_size - normal max_dis = data_set[1].max() for j in range(1000): threshold = max_dis / 1000 * j normal1 = 0 abnormal1 = 0 for k in range(data_set_size): if data_set[1][k] > threshold and data_set[2][k] == 1: normal1 += 1 if data_set[1][k] > threshold and data_set[2][k] == 2: abnormal1 += 1 roc_rate[0][j] = normal1 / normal # 阈值以上正常点/全体正常的点 roc_rate[1][j] = abnormal1 / abnormal # 阈值以上异常点/全体异常点 return roc_rate def test(training_filename, test_filename): # print("----hjz-----") print "----hjz-----" training_mat, training_label = file2mat(training_filename, 32) test_mat, test_label = file2mat(test_filename, 32) test_size = test_mat.shape[0] result = np.zeros((test_size, 3)) for i in range(test_size): print i result[i] = i + 1, classify(test_mat[i], training_mat), test_label[i] # 序号， 最小欧氏距离， 测试集数据类别 print result[i] result = np.transpose(result) # 矩阵转置 plt.figure(1) plt.scatter(result[0], result[1], c=result[2], edgecolors='None', s=1, alpha=1) # 图1 散点图：横轴为序号，纵轴为最小欧氏距离，点中心颜色根据测试集数据类别而定， 点外围无颜色，点大小为最小1，灰度为最大1 roc_rate = roc(result) plt.figure(2) plt.scatter(roc_rate[0], roc_rate[1], edgecolors='None', s=1, alpha=1) # 图2 ROC曲线， 横轴误报率，即阈值以上正常点/全体正常的点；纵轴检测率，即阈值以上异常点/全体异常点 plt.show() if __name__ == "__main__": print "----hjz-----" test('training.csv', 'test.csv')

Answer 1

这个图片输出，咋知道有啥意义？ # coding=utf-8 from __future__ import division import numpy as np import matplotlib.pyplot as plt def classify(input_vct, data_set): data_set_size = data_set.shape[0] diff_mat = np.tile(input_vct, (data_set_size, 1)) - data_set # 扩充input_vct到与data_set同型并相减 sq_diff_mat = diff_mat**2 # 矩阵中每个元素都平方 distance = sq_diff_mat.sum(axis=1)**0.5 # 每行相加求和并开平方根 return distance.min(axis=0) # 返回最小距离 def file2mat(test_filename, para_num): """ 将表格存入矩阵，test_filename为表格路径，para_num为存入矩阵的列数 返回目标矩阵，和矩阵每一行数据的类别 """ fr = open(test_filename) lines = fr.readlines() line_nums = len(lines) result_mat = np.zeros((line_nums, para_num)) # 创建line_nums行，para_num列的矩阵 class_label = [] for i in range(line_nums): line = lines[i].strip() item_mat = line.split(',') result_mat[i, :] = item_mat[0: para_num] class_label.append(item_mat[-1]) # 表格中最后一列正常1异常2的分类存入class_label fr.close() return result_mat, class_label def roc(data_set): normal = 0 data_set_size = data_set.shape[1] roc_rate = np.zeros((2, data_set_size)) for i in range(data_set_size): if data_set[2][i] == 1: normal += 1 abnormal = data_set_size - normal max_dis = data_set[1].max() for j in range(1000): threshold = max_dis / 1000 * j normal1 = 0 abnormal1 = 0 for k in range(data_set_size): if data_set[1][k] > threshold and data_set[2][k] == 1: normal1 += 1 if data_set[1][k] > threshold and data_set[2][k] == 2: abnormal1 += 1 roc_rate[0][j] = normal1 / normal # 阈值以上正常点/全体正常的点 roc_rate[1][j] = abnormal1 / abnormal # 阈值以上异常点/全体异常点 return roc_rate def test(training_filename, test_filename): # print("----hjz-----") print "----hjz-----" training_mat, training_label = file2mat(training_filename, 32) test_mat, test_label = file2mat(test_filename, 32) test_size = test_mat.shape[0] result = np.zeros((test_size, 3)) for i in range(test_size): print i result[i] = i + 1, classify(test_mat[i], training_mat), test_label[i] # 序号， 最小欧氏距离， 测试集数据类别 print result[i] result = np.transpose(result) # 矩阵转置 plt.figure(1) plt.scatter(result[0], result[1], c=result[2], edgecolors='None', s=1, alpha=1) # 图1 散点图：横轴为序号，纵轴为最小欧氏距离，点中心颜色根据测试集数据类别而定， 点外围无颜色，点大小为最小1，灰度为最大1 roc_rate = roc(result) plt.figure(2) plt.scatter(roc_rate[0], roc_rate[1], edgecolors='None', s=1, alpha=1) # 图2 ROC曲线， 横轴误报率，即阈值以上正常点/全体正常的点；纵轴检测率，即阈值以上异常点/全体异常点 plt.show() if __name__ == "__main__": print "----hjz-----" test('training.csv', 'test.csv')