XGBoost 算法使用示例

xgboost

使用 XGBoost 算法对泰坦尼克号乘客生还情况进行分类预测，并使用网格搜索法对算法的超参数进行调优，从而提高预测性能。

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from xgboost import XGBClassifier
from sklearn.model_selection import train_test_split
from sklearn.feature_extraction import DictVectorizer
from sklearn.tree import DecisionTreeClassifier,export_graphviz

#获取数据
data = pd.read_csv("./taiTanNiKeHao.csv")
#基本数据处理
#确定特征值,目标值
x = data[["Pclass","Age","Sex"]]
y = data['Survived']
#缺失值处理
x['Age'].fillna(value=data["Age"].mean(),inplace=True)
#分割数据
x_train,x_test,y_train,y_test = train_test_split(x,y,random_state=22,test_size=0.2)
#特征工程(字典特征提取)
x_train = x_train.to_dict(orient="records")
x_test = x_test.to_dict(orient="records")
transfer = DictVectorizer()
x_train = transfer.fit_transform(x_train)
x_test = transfer.fit_transform(x_test)

#模型训练
xg = XGBClassifier()
xg.fit(x_train,y_train)

#输出训练结果
ret = xg.score(x_test,y_test)
print("准确率为:\n",ret)

#对max_depth进行调优
depth_range = range(10)
score = []

for i in depth_range:
    xg = XGBClassifier(eta=1,gamma=0,max_depth=i)
    xg.fit(x_train,y_train)
    s = xg.score(x_test,y_test)
    print(s)
    score.append(s)

#调优结果可视化
plt.plot(depth_range,score)
plt.show()