knn_test.py

import numpy as np 
import pandas as pd 
import matplotlib.pyplot as plt


data = pd.read_csv('http_final_revise_test.csv', encoding="euc-kr")

df_z = np.array(data.dip)
df_xy = data[["sport","dport"]]
data_xy = data[["sport","dport"]].values

s_port = int(input('sport : '))
d_port = int(input('dport : '))
target = [s_port, d_port]

def data_set():
    size = len(df_xy)
    class_target = np.tile(target, (size,1))
    class_z = np.array(df_z)
    return df_xy, class_target, class_z
dataset, class_target, class_z = data_set()

def classify(dataset, class_target, class_category, k): 
    diffMat = class_target - dataset 
    sqDiffMat = diffMat**2 
    row_sum = sqDiffMat.sum(axis=1) 
    distance = np.sqrt(row_sum) 
    sortDist = distance.argsort() 
    
    class_result = {} 
    for i in range(k): 
        c = class_category[sortDist[i]] 
        class_result[c] = class_result.get(c, 0) + 1 
        return class_result


k = int(input('k값을 입력해주세요 :')) 
class_result = classify(data_xy, class_target, class_z, k) # classify()함수호출 
print(class_result)


def resultprint(class_result): 
    a_1 = a_2 = a_3 = a_4 = a_5 = 0
    
    for c in class_result.keys(): 
        if c == '172.30.1.53': 
            a_1 = class_result[c] 
        elif c == '210.102.100.235': 
            a_2 = class_result[c] 
        elif c == '218.54.130.120':  
            a_3 = class_result[c] 
        elif c == '218.54.130.30': 
            a_4 = class_result[c] 
        else : 
            a_5 = class_result[c] 
        
    if a_1 > a_2 and a_1 > a_3 and a_1 > a_4 and a_1 > a_5: 
        result = "a_1" 
    elif a_2 > a_1 and a_2 > a_3 and a_2 > a_4 and a_2 > a_5:
        result = "a_2" 
    elif a_3 > a_2 and a_3 > a_1 and a_3 > a_4 and a_3 > a_5:
        result = "a_3" 
    elif a_4 > a_2 and a_4 > a_3 and a_4 > a_1 and a_4 > a_5:
        result = "a_4" 
    elif a_5 > a_2 and a_5 > a_3 and a_5 > a_4 and a_5 > a_1:
        result = "a_5" 
    else: 
        result = "k값을 변경해주세요." 
    return result


for c in range(len(df_xy)): 
    data_xy[c][1] = data_xy[c][1]/10 
    if df_z[c] == 'a_1': plt.scatter(data_xy[c][0],data_xy[c][1],marker='o',color='b') 
    elif df_z[c] == 'a_2': plt.scatter(data_xy[c][0],data_xy[c][1],marker='o',color='g') 
    elif df_z[c] == 'a_3': plt.scatter(data_xy[c][0],data_xy[c][1],marker='o',color='y') 
    elif df_z[c] == 'a_4': plt.scatter(data_xy[c][0],data_xy[c][1],marker='o',color='c') 
    elif df_z[c] == 'a_5': plt.scatter(data_xy[c][0],data_xy[c][1],marker='o',color='m') 
    else: plt.scatter(data_xy[c][0],data_xy[c][1],marker='x',color='k') 
    plt.scatter(s_port,d_port,marker='*',color='r') 
    plt.show()