决策树实践

知识点补充
证明：

函数可以看成n 个优化函数之和

2：CART 分类算法

3：CART 回归算法

决策树算法代码实践
实验一：sklearn自带鸢尾花数据决策树分类
基本流程：
1：数据导入 from sklearn.data import load_iris
2：数据预处理 3类花4中特征 特征两两一组分开
3：每一组做一次并画出分类结果图，对某一组画出决策树分支图

import numpy as np
import pandas as pd
import matplotlib as mpl
import matplotlib.pyplot as plt
from sklearn.tree import DecisionTreeClassifier,plot_tree
from sklearn.preprocessing import LabelEncoder
from sklearn.metrics import accuracy_score
from sklearn.model_selection import train_test_split


if __name__=='__main__':
    mpl.rcParams['font.sans-serif'] = ['SimHei']
    mpl.rcParams['axes.unicode_minus'] = False
    iris_feature='花萼长度','花萼宽度','花瓣长度','花瓣宽度'
    path='..\\..\\..\\..\\..\\iris.data'
    data=pd.read_csv(path,header=None)
    x=data[list(range(4))]
    y=LabelEncoder().fit_transform(data[4])
    x_train_pre,x_test_pre,y_train,y_test = train_test_split(x,y,train_size=0.7,random_state=0)
    plt.figure(figsize=(6,6),facecolor='w')
    for Indexpair,pair in enumerate([[0,1],[0,2],[0,2],[1,2],[1,3],[2,3]]):
        x_train = x_train_pre[pair]
        x_test = x_test_pre[pair]
        model=DecisionTreeClassifier(max_depth=6,min_samples_leaf=3,criterion='entropy')
        model=model.fit(x_train,y_train)
        # 画图
        N,M=500,500
        x1_min,x2_min=x_train.min()
        x1_max, x2_max = x_train.max()
        t1=np.linspace(x1_min,x1_max,N)
        t2 = np.linspace(x2_min, x2_max, N)
        # 生成网格采样点
        x1,x2=np.meshgrid(t1,t2)