复习用scikit-learn和pandas进行线性回归

#导入的库
import matplotlib.pyplot as plt
%matplotlib inline
import numpy as np
import pandas as pd
from sklearn import datasets, linear_model

#读数据
data = pd.read_csv("Folds.csv")
data.head()

#前四列当作训练集
X = data[['AT','V','AP','RH']]
X.head()
#最后一列当作测试集
y = data[['PE']]
y.head()

#划分训练集和测试集
from sklearn.cross_validation import train_test_split
X_train,X_test,y_train,y_test = train_test_split(X,y,random_state=1)

#用scikit-learn的线性模型
from sklearn.linear_model import LinearRegression
train = LinearRegression()
train.fit(X_train,y_train)

'''
其目标函数为：g(x) = w1x1 + w2x2 + w3x3 + w4x4 + w0
coef_为w1到w4
intercept_为w0
'''
print(train.intercept_)
print(train.coef_)

#模型训练
y_pred = train.predict(X_test)

#模型评价，用scikit-learn计算MSE///计算RMSE
from sklearn import metrics
# 用scikit-learn计算MSE
print("MSE:",metrics.mean_squared_error(y_pred,y_test))
# 用scikit-learn计算RMSE
print("RMSE:",np.sqrt(metrics.mean_squared_error(y_test,y_pred)))

#用10折交叉验证
from sklearn.model_selection import cross_val_predict
predicted = cross_val_predict(train,X,y,cv = 10)
print("MSE:",metrics.mean_squared_error(y,predicted))
print("RMSE:",np.sqrt(metrics.mean_squared_error(y,predicted)))

#画图观察拟合情况
fig,ax = plt.subplots()
ax.scatter(y,predicted)
ax.plot([y.min(),y.max()],[y.min(),y.max()],'k--',lw=4)
ax.set_xlabel("Measured")
ax.set_ylabel('Predicted')
plt.show()

最后结果：