决策树

通过信息量计算

判断一个人是否买电脑的例子，建立决策树。

import numpy as np
import matplotlib.pyplot as plt
from sklearn.datasets import load_iris
from sklearn import preprocessing
from sklearn import tree
from sklearn.feature_extraction  import DictVectorizer
import csv

def read_data(filepath):
    Dtree = open(filepath,'r')
    reader= csv.reader(Dtree)
    headers = reader.__next__()
    list_data=[]
    list_label=[]
    for r in reader:
        dict_row={}
        for i in range(1,len(headers)-1):
            dict_row[headers[i]]=r[i]
        list_data.append(dict_row)
        list_label.append(r[-1])
    return list_data,list_label

def trans_data(data):
    vec = DictVectorizer()
    data = vec.fit_transform(data).toarray()
    vec_name = (vec.get_feature_names())
    return (data,vec_name)

def trans_label(data):
    lb = preprocessing.LabelBinarizer()
    data = lb.fit_transform(data)
    lb_name=lb.classes_
    return (data,lb_name)  

def main():
    print ("start".center(10,"-"))
    
    filepath= "data.csv"
    print ("ing".center(10,"-"))
    data,label =read_data(filepath)
    x_data,vec_name = trans_data(data)
    y_data,lb_name = trans_label(label)

    model = tree.DecisionTreeClassifier(criterion="entropy")
    model.fit(x_data,y_data)


    test = x_data[0]
    test_data = (test[np.newaxis,:])
    #test_data = test.reshape(1,-1)
    print("test:",test_data)
    predict = model.predict(test_data)
    print("result:",predict)
    
       
    print ("end".center(10,"-"))

    import graphviz
    dot_data = tree.export_graphviz(model,out_file=None,feature_names = vec_name,class_names=lb_name,filled=True,rounded=True,special_characters=True)
    graph = graphviz.Source(dot_data)
    graph.render("comuter")
main()

说明：首先，安装模块graphviz， pip intstall graphviz
其次，下载graphviz，https://www.graphviz.org/download/
之后，配置环境变量，将bin目录配置到系统path环境变量里.
最后，进入cmd，dot -c ,dot -v 测试。
成功后，运行代码即可。

通过基尼指数

import numpy as np
import matplotlib.pyplot as plt
from sklearn.datasets import load_iris
from sklearn import preprocessing
from sklearn import tree
from sklearn.feature_extraction  import DictVectorizer
import csv

def read_data(filepath):
    data = np.genfromtxt(filepath,delimiter=',')
    x_data =data[1:,1:-1]
    y_data =data[1:,-1]

    return x_data,y_data


def main():
    print ("start".center(10,"-"))
    
    filepath= "cart.csv"
    print ("ing".center(10,"-"))
    x_data,y_data =read_data(filepath)
    print ("x_data={}\ny_data={}.".format(x_data,y_data))
    model = tree.DecisionTreeClassifier()#基尼系数
    model.fit(x_data,y_data)

    import graphviz
    dot_data = tree.export_graphviz(model,out_file=None,feature_names = ["house_yes","house_no","single","married","divorced","income"],class_names=["no","yes"],filled=True,rounded=True,special_characters=True)
    graph = graphviz.Source(dot_data)
    graph.render("cart")
    print ("end".center(10,"-"))
main()