#?-*-?coding:?utf-8?-*-
“““
Created?on?Fri?Jul?10?22:04:33?2015

@author:?wepon
“““

import?numpy?as?np

class?DecisionTree:
????“““決策樹使用方法：
????
????????-?生成實例：?clf?=?DecisionTrees（）.?參數mode可選，ID3或C4.5，默認C4.5
????????
????????-?訓練，調用fit方法：?clf.fit（Xy）.??Xy均為np.ndarray類型
????????????????????????????
????????-?預測，調用predict方法：?clf.predict（X）.?X為np.ndarray類型
?????????????????????????????????
????????-?可視化決策樹，調用showTree方法?
????
????“““
????def?__init__（selfmode=‘C4.5‘）:
????????self._tree?=?None
????????
????????if?mode?==?‘C4.5‘?or?mode?==?‘ID3‘:
????????????self._mode?=?mode
????????else:
????????????raise?Exception（‘mode?should?be?C4.5?or?ID3‘）
????????
????????????
????
????def?_calcEntropy（selfy）:
????????“““
????????函數功能：計算熵
????????參數y：數據集的標簽
????????“““
????????num?=?y.shape[0]
????????#統計y中不同label值的個數，并用字典labelCounts存儲
????????labelCounts?=?{}
????????for?label?in?y:
????????????if?label?not?in?labelCounts.keys（）:?labelCounts[label]?=?0
????????????labelCounts[label]?+=?1
????????#計算熵
????????entropy?=?0.0
????????for?key?in?labelCounts:
????????????prob?=?float（labelCounts[key]）/num
????????????entropy?-=?prob?*?np.log2（prob）
????????return?entropy
????
????
????
????def?_splitDataSet（selfXyindexvalue）:
????????“““
????????函數功能：返回數據集中特征下標為index，特征值等于value的子數據集
????????“““
????????ret?=?[]
????????featVec?=?X[:index]
????????X?=?X[:[i?for?i?in?range（X.shape[1]）?if?i!=index]]
????????for?i?in?range（len（featVec））:
????????????if?featVec[i]==value:
????????????????ret.append（i）
????????return?X[ret:]y[ret]
????
????
????def?_chooseBestFeatureToSplit_ID3（selfXy）:
????????“““ID3
????????函數功能：對輸入的數據集，選擇最佳分割特征
????????參數dataSet：數據集，最后一列為label
????????主要變量說明：
????????????????numFeatures：特征個數
????????????????oldEntropy：原始數據集的熵
????????????????newEntropy：按某個特征分割數據集后的熵
????????????????infoGain：信息增益
????????????????bestInfoGain：記錄最大的信息增益
????????????????bestFeatureIndex：信息增益最大時，所選擇的分割特征的下標
????????“““
????????numFeatures?=?X.shape[1]
????????oldEntropy?=?self._calcEntropy（y）
????????bestInfoGain?=?0.0
????????bestFeatureIndex?=?-1
????????#對每個特征都計算一下infoGain，并用bestInfoGain記錄最大的那個
????????for?i?in?range（numFeatures）:????????
????????????featList?=?X[:i]
????????????uniqueVals?=?set（featList）
????????????newEntropy?=?0.0
????????????#對第i個特征的各個value，得到各個子數據集，計算各個子數據集的熵，
????????????#進一步地可以計算得到根據第i個特征分割原始數據集后的熵newEntropy
????????????for?value?in?uniqueVals:
????????????????sub_Xsub_y?=?self._splitDataSet（Xyivalue）
????????????????prob?=?len（sub_y）/float（len（y））
????????????????newEntropy?+=?prob?*?self._calcEntropy（sub_y）??
????????????#計算信息增益，根據信息增益選擇最佳分割特征
????????????infoGain?=?oldEntropy?-?newEntropy
????????????if?（infoGain?>?bestInfoGain）:
????????????????bestInfoGain?=?infoGain
????????????????bestFeatureIndex?=?i
????????return?bestFeatureIndex
????????
????def?_chooseBestFeatureToSplit_C45（selfXy）:
????????“““C4.5
????????????ID3算法計算的是信息增益，C4.5算法計算的是信息增益比，對上面ID3版本的函數稍作修改即可
????????“““
????????numFeatures?=?X.shape[1]
????????ol

?屬性????????????大小?????日期????時間???名稱
-----------?---------??----------?-----??----
?????目錄???????????0??2017-07-28?09:44??MachineLearning-master\
?????文件?????????180??2017-07-28?09:44??MachineLearning-master\.gitignore
?????目錄???????????0??2017-07-28?09:44??MachineLearning-master\DecisionTree\
?????文件????????9244??2017-07-28?09:44??MachineLearning-master\DecisionTree\id3_c45.py
?????文件????????1304??2017-07-28?09:44??MachineLearning-master\DecisionTree\readme.md
?????文件????????3283??2017-07-28?09:44??MachineLearning-master\DecisionTree\treePlotter.py
?????目錄???????????0??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\
?????目錄???????????0??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\cnn_LeNet\
?????文件???????12645??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\cnn_LeNet\convolutional_mlp.py
?????文件???????20706??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\cnn_LeNet\convolutional_mlp_commentate.py
?????目錄???????????0??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\dive_into_keras\
?????文件????????2197??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\dive_into_keras\cnn-svm.py
?????文件????????2550??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\dive_into_keras\cnn.py
?????文件?????????756??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\dive_into_keras\data.py
?????文件????????1432??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\dive_into_keras\get_feature_map.py
?????文件????????3300??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\dive_into_keras\README.md
?????目錄???????????0??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\FaceRecognition_CNN（olivettifaces）\
?????文件?????1182905??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\FaceRecognition_CNN（olivettifaces）\olivettifaces.gif
?????文件???????15554??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\FaceRecognition_CNN（olivettifaces）\train_CNN_olivettifaces.py
?????文件????????7042??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\FaceRecognition_CNN（olivettifaces）\use_CNN_olivettifaces.py
?????目錄???????????0??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\keras_usage\
?????文件????????5233??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\keras_usage\cnn.py
?????文件?????????764??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\keras_usage\data.py
?????文件????????9005??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\keras_usage\README.md
?????目錄???????????0??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\mlp\
?????文件???????14181??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\mlp\mlp.py
?????文件???????17794??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\mlp\mlp_with_commentate.py
?????文件?????????527??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\readme.md
?????目錄???????????0??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\Softmax_sgd（or?logistic_sgd）\
?????文件????????9456??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\Softmax_sgd（or?logistic_sgd）\logistic_sgd.py
?????文件???????19117??2017-07-28?09:44??MachineLearning-master\DeepLearning?Tutorials\Softmax_sgd（or?logistic_sgd）\logistic_sgd_commentate.py
............此處省略613個文件信息