2017-03-18_添加交流的课程注释

src/python/02.kNN/kNN.py

@@ -23,6 +23,8 @@ def createDataSet():
 
 def classify0(inX, dataSet, labels, k):
     """
+    inx[1,2,3]
+    DS=[[1,2,3],[1,2,0]]
     inX: 用于分类的输入向量
     dataSet: 输入的训练样本集
     labels: 标签向量
@@ -36,6 +38,10 @@ def classify0(inX, dataSet, labels, k):
     dataSetSize = dataSet.shape[0]
     # tile生成和训练样本对应的矩阵，并与训练样本求差
     diffMat = tile(inX, (dataSetSize, 1)) - dataSet
+    """
+    [[1,2,3],[1,2,3]]-[[1,2,3],[1,2,0]]
+    (A1-A2)^2+(B1-B2)^2+(c1-c2)^2
+    """
     # 取平方
     sqDiffMat = diffMat ** 2
     # 将矩阵的每一行相加
@@ -64,7 +70,7 @@ def test1():
     group, labels = createDataSet()
     print str(group)
     print str(labels)
-    print classify0([0, 0], group, labels, 3)
+    print classify0([0.1, 0.1], group, labels, 3)
 
 
 # ----------------------------------------------------------------------------------------
@@ -119,7 +125,7 @@ def datingClassTest():
     """
     hoRatio = 0.9  # 测试范围,一部分测试一部分作为样本
     # 从文件中加载数据
-    datingDataMat, datingLabels = file2matrix('../../../testData/datingTestSet2.txt')  # load data setfrom file
+    datingDataMat, datingLabels = file2matrix('testData/datingTestSet2.txt')  # load data setfrom file
     # 归一化数据
     normMat, ranges, minVals = autoNorm(datingDataMat)
     m = normMat.shape[0]
@@ -153,7 +159,7 @@ def img2vector(filename):
 def handwritingClassTest():
     # 1. 导入数据
     hwLabels = []
-    trainingFileList = listdir('../../../testData/trainingDigits')  # load the training set
+    trainingFileList = listdir('testData/trainingDigits')  # load the training set
     m = len(trainingFileList)
     trainingMat = zeros((m, 1024))
     for i in range(m):
@@ -161,17 +167,17 @@ def handwritingClassTest():
         fileStr = fileNameStr.split('.')[0]  # take off .txt
         classNumStr = int(fileStr.split('_')[0])
         hwLabels.append(classNumStr)
-        trainingMat[i, :] = img2vector('../../../testData/trainingDigits/%s' % fileNameStr)
+        trainingMat[i, :] = img2vector('testData/trainingDigits/%s' % fileNameStr)
 
     # 2. 导入测试数据
-    testFileList = listdir('../../../testData/testDigits')  # iterate through the test set
+    testFileList = listdir('testData/testDigits')  # iterate through the test set
     errorCount = 0.0
     mTest = len(testFileList)
     for i in range(mTest):
         fileNameStr = testFileList[i]
         fileStr = fileNameStr.split('.')[0]  # take off .txt
         classNumStr = int(fileStr.split('_')[0])
-        vectorUnderTest = img2vector('../../../testData/testDigits/%s' % fileNameStr)
+        vectorUnderTest = img2vector('testData/testDigits/%s' % fileNameStr)
         classifierResult = classify0(vectorUnderTest, trainingMat, hwLabels, 3)
         print "the classifier came back with: %d, the real answer is: %d" % (classifierResult, classNumStr)
         if (classifierResult != classNumStr): errorCount += 1.0

src/python/03.DecisionTree/DecisionTree.py

@@ -31,6 +31,7 @@ def createDataSet():
     #         ['no'],
     #         ['no'],
     #         ['no']]
+    # labels  露出水面   脚蹼
     labels = ['no surfacing', 'flippers']
     # change to discrete values
     return dataSet, labels
@@ -122,10 +123,9 @@ def chooseBestFeatureToSplit(dataSet):
             subDataSet = splitDataSet(dataSet, i, value)
             prob = len(subDataSet)/float(len(dataSet))
             newEntropy += prob * calcShannonEnt(subDataSet)
-        # gain[信息增益] 值越大，意味着该分类提供的信息量越大，该特征对分类的不确定程度越小
-        # 也就说： 列进行group分组后，对应的类别越多，信息量越大，那么香农熵越小，那么信息增益就越大，所以gain越大
+        # gain[信息增益]: 划分数据集前后的信息变化， 获取信息熵最大的值
         infoGain = baseEntropy - newEntropy
-        # print 'infoGain=', infoGain, 'bestFeature=', i
+        print 'infoGain=', infoGain, 'bestFeature=', i, baseEntropy, newEntropy
         if (infoGain > bestInfoGain):
             bestInfoGain = infoGain
             bestFeature = i
@@ -133,7 +133,7 @@ def chooseBestFeatureToSplit(dataSet):
 
 
 def majorityCnt(classList):
-    """majorityCnt(选择出线次数最多的一个结果)
+    """majorityCnt(选择出现次数最多的一个结果)
 
     Args:
         classList label列的集合

src/python/04.NaiveBayes/bayes.py

@@ -1,6 +1,10 @@
 #!/usr/bin/env python
 # -*- coding:utf-8 -*-
 from numpy import *
+"""
+p(xy)=p(x|y)p(y)=p(y|x)p(x)
+p(x|y)=p(y|x)p(x)/p(y)
+"""
 
 
 def loadDataSet():
@@ -8,7 +12,7 @@ def loadDataSet():
     创建数据集
     :return: 单词列表postingList, 所属类别classVec
     """
-    postingList = [['my', 'dog', 'has', 'flea', 'problems', 'help', 'please'],
+    postingList = [['my', 'dog', 'has', 'flea', 'problems', 'help', 'please'], #[0,0,1,1,1......]
                    ['maybe', 'not', 'take', 'him', 'to', 'dog', 'park', 'stupid'],
                    ['my', 'dalmation', 'is', 'so', 'cute', 'I', 'love', 'him'],
                    ['stop', 'posting', 'stupid', 'worthless', 'garbage'],
@@ -37,7 +41,7 @@ def setOfWords2Vec(vocabList, inputSet):
     :param inputSet: 输入数据集
     :return: 匹配列表[0,1,0,1...]
     """
-    returnVec = [0] * len(vocabList)
+    returnVec = [0] * len(vocabList)# [0,0......]
     for word in inputSet:
         if word in vocabList:
             returnVec[vocabList.index(word)] = 1
@@ -49,8 +53,8 @@ def setOfWords2Vec(vocabList, inputSet):
 def _trainNB0(trainMatrix, trainCategory):
     """
     训练数据原版
-    :param trainMatrix: 文件单词矩阵
-    :param trainCategory: 文件对应的类别
+    :param trainMatrix: 文件单词矩阵 [[1,0,1,1,1....],[],[]...]
+    :param trainCategory: 文件对应的类别[0,1,1,0....]
     :return:
     """
     # 文件数
@@ -60,21 +64,21 @@ def _trainNB0(trainMatrix, trainCategory):
     # 侮辱性文件的出现概率
     pAbusive = sum(trainCategory) / float(numTrainDocs)
     # 构造单词出现次数列表
-    p0Num = zeros(numWords)
-    p1Num = zeros(numWords)
+    p0Num = zeros(numWords)[0,0,0,.....]
+    p1Num = zeros(numWords)[0,0,0,.....]
 
     # 整个数据集单词出现总数
     p0Denom = 0.0
     p1Denom = 0.0
     for i in range(numTrainDocs):
         if trainCategory[i] == 1:
-            p1Num += trainMatrix[i]
+            p1Num += trainMatrix[i] #[0,1,1,....]->[0,1,1,...]
             p1Denom += sum(trainMatrix[i])
         else:
             p0Num += trainMatrix[i]
             p0Denom += sum(trainMatrix[i])
     # 类别1，即侮辱性文档的[P(F1|C1),P(F2|C1),P(F3|C1),P(F4|C1),P(F5|C1)....]列表
-    p1Vect = p1Num / p1Denom
+    p1Vect = p1Num / p1Denom# [1,2,3,5]/90->[1/90,...]
     # 类别0，即正常文档的[P(F1|C0),P(F2|C0),P(F3|C0),P(F4|C0),P(F5|C0)....]列表
     p0Vect = p0Num / p0Denom
     return p0Vect, p1Vect, pAbusive
@@ -94,7 +98,7 @@ def trainNB0(trainMatrix, trainCategory):
     # 侮辱性文件的出现概率
     pAbusive = sum(trainCategory) / float(numTrainDocs)
     # 构造单词出现次数列表
-    p0Num = ones(numWords)
+    p0Num = ones(numWords)#[0,0......]->[1,1,1,1,1.....]
     p1Num = ones(numWords)
 
     # 整个数据集单词出现总数，2.0根据样本/实际调查结果调整分母的值
@@ -117,7 +121,7 @@ def trainNB0(trainMatrix, trainCategory):
 def classifyNB(vec2Classify, p0Vec, p1Vec, pClass1):
     """
     使用算法
-    :param vec2Classify: 待测数据
+    :param vec2Classify: 待测数据[0,1,1,1,1...]
     :param p0Vec: 类别1，即侮辱性文档的[log(P(F1|C1)),log(P(F2|C1)),log(P(F3|C1)),log(P(F4|C1)),log(P(F5|C1))....]列表
     :param p1Vec: 类别0，即正常文档的[log(P(F1|C0)),log(P(F2|C0)),log(P(F3|C0)),log(P(F4|C0)),log(P(F5|C0))....]列表
     :param pClass1: 类别1，侮辱性文件的出现概率

src/python/12.FrequentPattemTree/fpGrowth.py

@@ -1,3 +1,6 @@
+#!/usr/bin/python
+# coding:utf8
+
 '''
 Created on Jun 14, 2011
 FP-Growth FP means frequent pattern
@@ -22,15 +25,41 @@ class treeNode:
         self.children = {}
 
     def inc(self, numOccur):
+        """inc(对count变量增加给定值)
+
+        """
         self.count += numOccur
 
     def disp(self, ind=1):
+        """disp(用于将树以文本形式显示)
+
+        """
         print '  '*ind, self.name, ' ', self.count
         for child in self.children.values():
             child.disp(ind+1)
 
 
+def loadSimpDat():
+    simpDat = [['r', 'z', 'h', 'j', 'p'],
+               ['z', 'y', 'x', 'w', 'v', 'u', 't', 's'],
+               ['z'],
+               ['r', 'x', 'n', 'o', 's'],
+               ['y', 'r', 'x', 'z', 'q', 't', 'p'],
+               ['y', 'z', 'x', 'e', 'q', 's', 't', 'm']]
+    return simpDat
+
+
+def createInitSet(dataSet):
+    retDict = {}
+    for trans in dataSet:
+        retDict[frozenset(trans)] = 1
+    return retDict
+
+
 def createTree(dataSet, minSup=1): #create FP-tree from dataset but don't mine
+    """
+
+    """
     headerTable = {}
     #go over dataSet twice
     for trans in dataSet:#first pass counts frequency of occurance
@@ -76,6 +105,21 @@ def updateHeader(nodeToTest, targetNode):   #this version does not use recursion
     nodeToTest.nodeLink = targetNode
 
 
+if __name__ == "__main__":
+    rootNode = treeNode('pyramid', 9, None)
+    rootNode.children['eye'] = treeNode('eye', 13, None)
+    rootNode.children['phoenix'] = treeNode('phoenix', 3, None)
+    # 将树以文本形式显示
+    # print rootNode.disp()
+
+    # load样本数据
+    simpDat = loadSimpDat()
+    print simpDat
+    # 重新装载 frozen set 格式化样本数据，用dist存储数据和对应的次数
+    initSet = createInitSet(simpDat)
+    print initSet
+
+
 def ascendTree(leafNode, prefixPath): #ascends from leaf node to root
     if leafNode.parent != None:
         prefixPath.append(leafNode.name)
@@ -111,21 +155,6 @@ def mineTree(inTree, headerTable, minSup, preFix, freqItemList):
             mineTree(myCondTree, myHead, minSup, newFreqSet, freqItemList)
 
 
-def loadSimpDat():
-    simpDat = [['r', 'z', 'h', 'j', 'p'],
-               ['z', 'y', 'x', 'w', 'v', 'u', 't', 's'],
-               ['z'],
-               ['r', 'x', 'n', 'o', 's'],
-               ['y', 'r', 'x', 'z', 'q', 't', 'p'],
-               ['y', 'z', 'x', 'e', 'q', 's', 't', 'm']]
-    return simpDat
-
-
-def createInitSet(dataSet):
-    retDict = {}
-    for trans in dataSet:
-        retDict[frozenset(trans)] = 1
-    return retDict
 
 
 import twitter