• 機器學習之分類決策樹DecisionTreeClassifier

# -*- coding: utf-8 -*-
"""
Created on Fri Nov 23 21:06:54 2018

@author: muli
"""

import numpy as np
from sklearn.tree import DecisionTreeClassifier
from sklearn import  datasets
from sklearn import cross_validation
import matplotlib.pyplot as plt


def load_data():
    '''
    載入用於分類問題的資料集。資料集採用 scikit-learn 自帶的 iris 資料集

    :return: 一個元組，用於分類問題。
    元組元素依次為：訓練樣本集、測試樣本集、訓練樣本集對應的標記、測試樣本集對應的標記
    '''
    # scikit-learn 自帶的 iris 資料集
    iris=datasets.load_iris() 
    X_train=iris.data
    y_train=iris.target
    # 分層取樣拆分成訓練集和測試集，測試集大小為原始資料集大小的 1/4
    return cross_validation.train_test_split(X_train, y_train,test_size=0.25,
		random_state=0,stratify=y_train)
    

def test_DecisionTreeClassifier(*data):
    '''
    測試 DecisionTreeClassifier 的用法

    :param data: 可變引數。
    它是一個元組，這裡要求其元素依次為：訓練樣本集、測試樣本集、訓練樣本的標記、測試樣本的標記
    :return:  None
    '''
    X_train,X_test,y_train,y_test=data
    clf = DecisionTreeClassifier()
    clf.fit(X_train, y_train)

    print("Training score:%f"%(clf.score(X_train,y_train)))
    print("Testing score:%f"%(clf.score(X_test,y_test)))
    

def test_DecisionTreeClassifier_criterion(*data):
    '''
    測試 DecisionTreeClassifier 的預測效能隨 criterion 引數的影響

    :param data:  可變引數。
    它是一個元組，這裡要求其元素依次為：訓練樣本集、測試樣本集、訓練樣本的標記、測試樣本的標記
    :return:  None
    '''
    X_train,X_test,y_train,y_test=data
    criterions=['gini','entropy']
    for criterion in criterions:
        clf = DecisionTreeClassifier(criterion=criterion)
        clf.fit(X_train, y_train)
        print("criterion:%s"%criterion)
        print("Training score:%f"%(clf.score(X_train,y_train)))
        print("Testing score:%f"%(clf.score(X_test,y_test)))
        print("-------")


def test_DecisionTreeClassifier_splitter(*data):
    '''
    隨劃分型別--最優劃分和隨機劃分

    :param data: 可變引數。
    它是一個元組，這裡要求其元素依次為：訓練樣本集、測試樣本集、訓練樣本的標記、測試樣本的標記
    :return:  None
    '''
    X_train,X_test,y_train,y_test=data
    splitters=['best','random']
    for splitter in splitters:
        clf = DecisionTreeClassifier(splitter=splitter)
        clf.fit(X_train, y_train)
        print("splitter:%s"%splitter)
        print("Training score:%f"%(clf.score(X_train,y_train)))
        print("Testing score:%f"%(clf.score(X_test,y_test)))
        print("------------------")


def test_DecisionTreeClassifier_depth(*data,maxdepth):
    '''
    測試預測效能隨 max_depth 引數的影響

    :param data:  可變引數。它是一個元組，這裡要求其元素依次為：訓練樣本集、測試樣本集、訓練樣本的標記、測試樣本的標記
    :param maxdepth: 一個整數，用於 DecisionTreeClassifier 的 max_depth 引數
    :return:  None
    '''
    X_train,X_test,y_train,y_test=data
    depths=np.arange(1,maxdepth)
    training_scores=[]
    testing_scores=[]
    for depth in depths:
        clf = DecisionTreeClassifier(max_depth=depth)
        clf.fit(X_train, y_train)
        training_scores.append(clf.score(X_train,y_train))
        testing_scores.append(clf.score(X_test,y_test))

    ## 繪圖
    fig=plt.figure()
    ax=fig.add_subplot(1,1,1)
    ax.plot(depths,training_scores,label="traing score",marker='o')
    ax.plot(depths,testing_scores,label="testing score",marker='*')
    ax.set_xlabel("maxdepth")
    ax.set_ylabel("score")
    ax.set_title("Decision Tree Classification")
    ax.legend(framealpha=0.5,loc='best')
    plt.show()




if __name__=='__main__':
    # 產生用於分類問題的資料集
    X_train,X_test,y_train,y_test=load_data() 
    # 呼叫 test_DecisionTreeClassifier
#    test_DecisionTreeClassifier(X_train,X_test,y_train,y_test)
    # 呼叫 test_DecisionTreeClassifier_criterion
#    test_DecisionTreeClassifier_criterion(X_train,X_test,y_train,y_test) 
    # 呼叫 test_DecisionTreeClassifier_splitter
#    test_DecisionTreeClassifier_splitter(X_train,X_test,y_train,y_test) 
    # 呼叫 test_DecisionTreeClassifier_depth
    test_DecisionTreeClassifier_depth(X_train,X_test,y_train,y_test,maxdepth=15)