k近邻算法

怎样计算两个数据的距离（相近程度）
假如两个数据有A B，有两个属性 p1 p2
那么，A与B的距离 distance = 开平方( 平方(A.p1 - B.p1)+ 平方(A.p2 - B.p2) )

import pandas as pd
import numpy as np
from sklearn import preprocessing
from sklearn.neighbors import KNeighborsRegressor

df = pd.read_csv("./titanic_train.csv")
#drop NAN datas
df = df.dropna()

#75%用来训练
trainCount = int(df.shape[0]*0.75)

df_train = df.iloc[:trainCount]
df_test = df.iloc[trainCount:]


#选择哪些列用来做比较
cols = ['Age','Pclass']
#初始化KNeighborsRegressor，指定取前10个最相似的近邻的平均值，默认是5个
knn = KNeighborsRegressor(n_neighbors=10)
#开始训练，目标是Fare的值
knn.fit(df_train[cols],df_train['Fare'])
#用测试集预测Fare值
predictions = knn.predict(df_test[cols])

print(predictions)
print(df_test['Fare'].values)

#评估测试集
from sklearn.metrics import mean_squared_error
mse = mean_squared_error(df_test['Fare'] , predictions)
rmse = mse ** (1/2)
print(rmse)