kNN

def myClassify0(inX, dataSet, labels, k) :
    distances = calculateEuclideanDistance(inX, dataSet)

    sortedDistIndices = distances.argsort()
    return findKClosestDataPoint(sortedDistIndices, labels, k)


def findKClosestDataPoint(sortedDistIndices, labels, k) :
    classCount = {}
    for i in range(k) :
        voteIlabel = labels[sortedDistIndices[i]]
        classCount[voteIlabel] = classCount.get(voteIlabel,0) + 1

    sortedClassCount = sorted(classCount.iteritems(), key = operator.itemgetter(1), reverse = True)

    return sortedClassCount[0][0]

def calculateEuclideanDistance(inX, dataSet) :
    diffMatrix = getDifferences(inX, dataSet)
    squareDiffMatrix = getSquareOfDifferences(diffMatrix)
    sqDistance = sumSqDifferences(squareDiffMatrix)
    return getSquareRootOfSum(sqDistance)

def getDifferences(inX, dataSet) :
    dataSetSize = dataSet.shape[0]
    diffMat = tile(inX, (dataSetSize, 1)) - dataSet

    return diffMat

def getSquareOfDifferences(sqDiffMat) :
    return sqDiffMat**2

def sumSqDifferences(sqDiffMat) :
    return sqDiffMat.sum(axis=1)

def getSquareRootOfSum(sqDistance) :
    return sqDistance ** 0.5