python机器学习决策树初识

运用python进行决策树分类预测。数据采用鸢尾花数据。（我也是python新手，将视频中的代码敲下来自练习消化。）

# -*- coding: utf-8 -*-
import numpy as np
import pandas as pd
from sklearn.datasets import  load_iris #导入鸢尾花数据
from sklearn import tree       #导入机器学习决策树包
from sklearn.cross_validation import train_test_split

def main():
    iris = load_iris()
    print iris
    print len(iris["data"])       #数据长度
    train_data,test_data,train_target,test_target = train_test_split(iris.data,iris.target,test_size=0.2,random_state=1)

#建模
    clf  = tree.DecisionTreeClassifier(criterion="entropy")   #分类器决策树 熵增益
    clf.fit(train_data,train_target)        #建立模型
    y_pred = clf.predict(test_data)        #测试集预测

#验证
    from sklearn import metrics
    print metrics.accuracy_score(y_true = test_target,y_pred = y_pred) #真实值和预测值  给出预测精度
    print metrics.confusion_matrix(y_true=test_target,y_pred = y_pred)  #混淆矩阵

    with open("./data/tree.dot","w") as fw:                 #将决策树输出值文件
        tree.export_graphviz(clf,out_file=fw)
if __name__ == "__main__":
    main()