1. 程式人生 > >機器學習—線性回歸

機器學習—線性回歸

com str mode imp repr 線性模型 images mage 訓練集

一、普通的線性模型

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.linear_model import LinearRegression
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn import metrics
%matplotlib inline
data = pd.read_csv(
Advertising.csv,index_col=0)#第一列為index data.head()

技術分享

#切分訓練集和測試集
x = data.values[:,:3]
y = data.values[:,3]
x_train,x_test,y_train,y_test = train_test_split(x,y,train_size=0.7,random_state=0)
#標準化處理
sc = StandardScaler()
x_train_std = sc.fit_transform(x_train)
x_test_std = sc.transform(x_test)
#訓練模型
linreg = LinearRegression()
linreg.fit(x_train_std,y_train)
y_pred 
= linreg.predict(x_test_std) #檢驗模型結果 mse = np.average((y_pred-y_test)**2) metrics.mean_squared_error(y_pred,y_test) #這個也是均方誤差 r2 = metrics.r2_score(y_test,y_pred) #R2值,註意參數,前面的是實際值,後面的是預測值 mse,r2 #計算R2 def calculater2(y_pred,y_test): RSS = ((y_pred-y_test)**2).sum() TSS = (((y_test-np.average(y_test))**2)).sum()
return 1-(RSS/TSS) calculater2(y_pred,y_test) #畫圖 fig = plt.figure(figsize=(10,6)) plt.plot(y_test) plt.plot(y_pred)

技術分享

二、加入正則化的模型

Ridge回歸

from sklearn.linear_model import RidgeCV,LassoCV   #用這個自帶交叉驗證參數
from sklearn.model_selection import GridSearchCV   #如果使用RidgeCV就不用GridSearchCV這個API了
#使用RidgeCV來建立參數
alpha = np.logspace(-3,2,10)    #生成超參數,10的-3次方到10的2次方的等差數列
ridge = RidgeCV(alpha,cv=5)
ridge.fit(x_train_std,y_train)
ridge.alpha_   #輸出超參數的值
#使用Ridge配合GridSearchCV來做
from sklearn.linear_model import Ridge,Lasso
ridge_model = GridSearchCV(Ridge(),param_grid={alpha:alpha},cv=5)
ridge_model.fit(x_train_std,y_train)
ridge_model.best_params_
#驗證模型效果
y_pred_ridge = ridge.predict(x_test_std)
mse_ridge = metrics.mean_squared_error(y_test,y_pred_ridge)
r2_ridge = metrics.r2_score(y_test,y_pred_ridge)
mse_ridge,r2_ridge

Lasso回歸

#建立模型
lasso = LassoCV(alphas=alpha,cv=5)
lasso.fit(x_train_std,y_train)
lasso.alpha_
#驗證模型效果
y_pred_lasso = lasso.predict(x_test_std)
mse_lasso = metrics.mean_squared_error(y_test,y_pred_lasso)
r2_lasso = metrics.r2_score(y_test,y_pred_lasso)
mse_lasso,r2_lasso



機器學習—線性回歸