import numpy as np

import pandas as pd

import matplotlib.pyplot as plt

data = pd.read_excel('jiemo.xls')  # 导入当前文件夹下的58.csv数据文件

df=pd.DataFrame(data)

df1=df.dropna()

df1['平方']=df1['平方'].str.replace("㎡","")

df1['价格/每平方米']=df1['价格']/df1['平方'].astype(int)

d1=df1['地址']

category = pd.Categorical(d1)

from sklearn.preprocessing import LabelEncoder

le = LabelEncoder()

le.fit(category)

d1=le.transform(d1)

price = df1['价格/每平方米']

adress = d1

plt.scatter(price, adress)

# x，y取值范围设置

# 可以过滤掉一部分脏数据

# 设置title和x，y轴的label

plt.xlabel("price")

plt.ylabel("adress")

plt.show()

d1=df1['地址']

category = pd.Categorical(d1)

from sklearn.preprocessing import LabelEncoder

le = LabelEncoder()

le.fit(category)

d1=le.transform(d1)

print(d1)

d2=df1['朝向']

category = pd.Categorical(d2)

le.fit(category)

d2=le.transform(d2)

print(d2)

d3=df1['类型']

category = pd.Categorical(d3)

le.fit(category)

d3=le.transform(d3)

print(d3)

target=[]

t2=[]

feature=zip(d1,df1['平方'].astype(int),d2,d3)

for i in df1['价格'].astype(int):

    target.append(i)

for i in feature:

    t2.append(i)

[186 163 187 ... 235 231 231]

[31 33 31 ... 33 31 33]

[17 17  5 ... 17  2 39]

from sklearn.neighbors import KNeighborsClassifier

knn = KNeighborsClassifier(n_neighbors=1)

knn.fit(t2,target)

knn.score(t2,target)

knn.predict([[167,80,2,2]])