pbharrin / machinelearninginaction3x Goto Github PK

In chapter 11

def aprioriGen(Lk, k): #creates Ck
    retList = []
    lenLk = len(Lk)
    for i in range(lenLk):
        for j in range(i+1, lenLk): 
            L1 = list(Lk[i])[:k-2]; L2 = list(Lk[j])[:k-2]
            L1.sort(); L2.sort()
            if L1==L2: #if first k-2 elements are equal
                retList.append(Lk[i] | Lk[j]) #set union
    return retList

I think the L1 and L2 should sort first, then extract the front k-2 items to compare. Take the example:
L1 = [1,2]
L2 = {3, 1}
Though, when a set converted to a list, it's usually sorted.

I'm getting an error with the classify0() function in the kNN.py module.

  9     for i in range(k):
 10         voteIlabel = labels[sortedDistIndicies[i]]

---> 11 classCount[voteIlabel] = classCount.get(voteIlabel, 0) + 1
12 sortedClassCount = sorted(classCount.items(), key=operator.itemgetter(1), reverse=True)
13 return sortedClassCount[0][0]

TypeError: unhashable type: 'list'

I've taken the code from the pbharrin/machinelearninginaction3x github repo so it shouldn't be down to typos. I've run the code line by line outside of the function and I get the same error at the same point when I try to run the for loop. I'm not an experienced Python coder so it's a bit beyond me to try and solve this myself. I'm loving the book so far though.

When we attempt to classify a document, we multiply a lot of probabilities together to
get the probability that a document belongs to a given class. This will look something
like p(w0|1)p(w1|1)p(w2|1). If any of these numbers are 0, then when we multiply
them together we get 0. To lessen the impact of this, we’ll initialize all of our occurrence counts to 1, and we’ll initialize the denominators to 2

First of all, huge thanks for creating an understandable ML book.

Coming back to question now:

I perfectly, understood the idea behind setting p0Num and p1Num to np.ones, but why you have set p0Denom and p1Denom both to 2?

line 99
del(list(trainingSet)[randIndex])
seems that the deletion does not work
maybe change to:
del(trainingSet[randIndex])

I read the book "Machine Learning in action" and repeat code from the chapter 2 kNN, but i got error
"ValueError: could not convert string to float:" on the string

In chapter 11

def generateRules(L, supportData, minConf=0.7):  #supportData is a dict coming from scanD
    bigRuleList = []
    for i in range(1, len(L)):#only get the sets with two or more items
        for freqSet in L[i]:
            H1 = [frozenset([item]) for item in freqSet]
            if (i > 1):
                rulesFromConseq(freqSet, H1, supportData, bigRuleList, minConf)
            else:
                calcConf(freqSet, H1, supportData, bigRuleList, minConf)
    return bigRuleList

When the frequent set is like {1,2,3}, it can't generate the rules like {1,2}->3.
Shoule the snippet

            if (i > 1):
                rulesFromConseq(freqSet, H1, supportData, bigRuleList, minConf)
            else:
                calcConf(freqSet, H1, supportData, bigRuleList, minConf)

change to

            rulesFromConseq(freqSet, H1, supportData, bigRuleList, minConf)
            if (i > 1):
                calcConf(freqSet, H1, supportData, bigRuleList, minConf)

in kMeans.py in CH10. it does not work in python3.

    def randCent(dataSet, k):
        n = shape(dataSet)[1]
        centroids = mat(zeros((k, n)))
        for j in range(n):
            minJ = min(dataSet[:,j])
            rangeJ = float(max(dataSet[:,j]) - minJ)

i got a error

rangeJ = float(max(dataSet[:,j]) - minJ)
TypeError: unsupported operand type(s) for -: 'map' and 'map'

Steps to reproduce:

1. cd machinelearninginaction3x
2. python3 -m venv venv
3. pip install numpy && pip install matplotlib
4. . venv/bin/activate
5. cd Ch02
6. python kNNTest.py
7. Got error

Traceback (most recent call last):
File "/usr/local/src/github/machinelearninginaction3x/Ch02/kNNTest.py", line 10, in
from Ch02 import kNN
ModuleNotFoundError: No module named 'Ch02'

When I changed "from Ch02 import kNN" to "import kNN", it works.

It seems that when you in the Ch02 directory, you should import a module directly.

def createInitSet(dataSet):
    retDict = {}
    for trans in dataSet:
        retDict[frozenset(trans)] = 1
    return retDict

the code retDict[frozenset(trans)] = 1, not consider when two transactions are the same.
should changed to

def createInitSet(dataSet):
    retDict = {}
    for trans in dataSet:
        retDict[frozenset(trans)] = retDict.get(frozenset(trans), 0) + 1
    return retDict