基于Bagging与boosting的算法4---XGBoost的应用

参考xgboost官网

xgboost官网查看公式推理。（难）
还有一个特征：xgboost可以在一定程度上进行并行

了解Adaboost-参数：

import pandas as pd 
from sklearn.ensemble import AdaBoostClassifier
from sklearn.model_selection import GridSearchCV,train_test_split
import matplotlib.pyplot as plt 

df =pd.read_csv("car.csv",encoding="gbk")
x_train,x_test,y_train,y_test=train_test_split(df.values[:,:-1],df.values[:,-1],test_size=0.3,random_state=10)

def get_n_estimators():
    estimators=range(5,100)
    score=[]
    for n_estimators in estimators:
        model=AdaBoostClassifier(n_estimators=n_estimators)
        model.fit(x_train,y_train)
        score.append(model.score(x_test,y_test))
    plt.plot(estimators,score)
    plt.show()

def get_learning_rate():
    learning_rate=[0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1,2,3]
    score=[]
    for i in learning_rate:
        model=AdaBoostClassifier(learning_rate=i)
        model.fit(x_train,y_train)
        score.append(model.score(x_test,y_test))
    plt.plot(learning_rate,score)
    plt.show()
    
if __name__=="__main__":
    get_n_estimators()

xgboost通用参数

nthread----多线程控制
8和9两个参数经常需要调整
一般调整的参数是：max_depth; eta;
silent;
colsample_bytree;
colsample_bylevel