机器学习实战 k近邻算法

import pandas as pd
import numpy as np
rawdata ={'电影名':['无问东西','后来的我们','前任3','红海行动','唐人街探案','战狼2'],
          '打斗次数':[1,5,12,108,112,115],
          '接吻镜头':[101,89,97,5,9,8],
          '电影类型':['爱情片','爱情片','爱情片','动作片','动作片','动作片']}
def classfy(rawdata,newdata,k):
    pandasdata = pd.DataFrame(rawdata)
    pandasdata.insert(0,pandasdata.pop('电影名'))   
    dist = list((((pdata.iloc[:, 1:3] - newdata) ** 2).sum(1)) ** 0.5)

    ord = np.argsort(dist)

    result = {}
    for i in range(k):
        if pandasdata['电影类型'][ord[i]] in result:
            result[pandasdata['电影类型'][ord[i]]] += 1
        result[pandasdata['电影类型'][ord[i]]] = 0
    return max(result, key=result.get)

def classfy1(rawdata,newdata,k):
    pandasdata = pd.DataFrame(rawdata)
    pandasdata.insert(0,pandasdata.pop('电影名'))  

    dist = list((((pdata.iloc[:, 1:3] - newdata) ** 2).sum(1)) ** 0.5)

    dr = pd.DataFrame({'dst': dist, 'label': pandasdata['电影类型']})
    dr1 = dr.sort_values(by='dst')[:k]
    re = dr1.loc[:, 'label'].value_counts()
    return re.index[0]

newdata = [24, 76]
k = 4
#re = classfy1(rawdata,newdata,k)
re = classfy1(rawdata,newdata,k)

print(re)

rawdata = pd.read_table(r'datingTestSet.txt',header=None)

import matplotlib
import matplotlib.pyplot as plt

color = []
for i in range(rawdata.shape[0]):
    label = rawdata.iloc[i,-1]
    if label == 'largeDoses':
        color.append('red')
    if label == 'smallDoses':
        color.append('orange')
    if label == 'didntLike':
        color.append('black')
print(len(color))
fig = plt.figure()
# fig.add_subplot(221)
plt.rcParams['font.sans-serif'] = ['Simhei']  #'FangSong'
plt.scatter(rawdata.iloc[:,1],rawdata.iloc[:,2],marker='.',c=color)
plt.xlabel("玩游戏视频所占时间比")
plt.ylabel("每周消费冰激凌公升数")
fig.show()