机器学习metrics模型评估

使用metrics对创建的模型进行评估。

 

import pandas as pd

 

from sklearn.linear_model import LinearRegression

import matplotlib.pyplot as plt

from sklearn.metrics import *

data1=pd.read_csv('data.csv')

print(data1)

data1.dropna(axis=0,how='any',inplace=True)

data1['总价']=data1['总价'].map(lambda z: z.replace('万',''))

data1['建筑面积']=data1['建筑面积'].map(lambda p: p.replace('平米',''))

data1['总价']=data1['总价'].astype(float)

data1['建筑面积']=data1['建筑面积'].astype(float)

x=pd.DataFrame(data1['总价'])

y=pd.DataFrame(data1['建筑面积'])

clf=LinearRegression()

clf.fit(x,y)

w=clf.coef_

b=clf.intercept_

x1=180

y1=x1*w+b

print(y1)

print(clf.score(x, y)*100)

y_pred=clf.predict(x)

s1=explained_variance_score(x, y)

print('回归方差为：\n',s1)

s2=mean_absolute_error(x, y)

print('平均绝对误差为：\n',s2)

s3=mean_squared_error(x,y)

print('均方差为：\n',s3)

s4=median_absolute_error(x, y)

print('中值绝对误差为：\n',s4)

s5=r2_score(x, y)

print('R平方值为：\n',s5)