机器学习实战：决策树-隐形眼镜

import pandas as pd
from sklearn.preprocessing import LabelEncoder,OneHotEncoder
from sklearn.tree import DecisionTreeClassifier,export_graphviz
import pydotplus
import numpy as np
import graphviz


"""
函数说明：数据处理
Parameters:
    filename - 数据文件
Returns:
    
"""
def dataDeal(filename):
    with open(filename,'r') as fr:
        lenses = [ins.strip().split('\t') for ins in fr.readlines()]
    lenses_target = []
    lincelist = []
    linceDict = {}
    for each in lenses:
        lenses_target.append(each[-1])  # 目标
    lensesLabels = ['age', 'prescript', 'astigmatic', 'tearRate']  # 特征标签
    for label in lensesLabels:
        for each in lenses:
            lincelist.append(each[lensesLabels.index(label)])
        linceDict[label] = lincelist
        lincelist = []
    lince_pd = pd.DataFrame(linceDict)  # 将数据转化为DataFrame格式
    le = LabelEncoder()
    for col in lince_pd.columns:
        lince_pd[col] = le.fit_transform(lince_pd[col])  # 数据序列化(对数据进行编码)
    return lince_pd,lenses_target

if __name__ == '__main__':
    lince_pd,lenses_target= dataDeal('lenses.txt')
    clf = DecisionTreeClassifier(max_depth=4)
    clf = clf.fit(lince_pd.values.tolist(),lenses_target)
    print(clf.predict([[1,1,1,0]]))  # 预测
    lensesLabels = ['age', 'prescript', 'astigmatic', 'tearRate']  # 特征标签
    dot_data = export_graphviz(clf, feature_names=lensesLabels, out_file=None)
    dot = graphviz.Source(dot_data)
    dot.view()