Machine learning Combat--the second chapter of improving the pairing effect of dating sites

Three types: dislike-1, charm-2, attractive-3.

Sample characteristics: The number of frequent flyer miles earned per year, the percentage of time spent playing video games, and the number of ice cream litres consumed per week.

1  fromNumPyImport*2 Importmatplotlib3 ImportMatplotlib.pyplot as Plt4 5 " "6 input text file name string, output training sample matrix and class label vector7 " "8 defFile2matrix (filename):9FR =open (filename)TenArrayolines =fr.readlines () #一次读取整个文件, parses the contents of a file into a list of rows  OneNumberOfLines =Len (arrayolines) AReturnmat = Zeros ((NumberOfLines, 3)) -Classlabelvector = [] -index =0 the      forLineinchArrayolines: -line = Line.strip ()#intercept all carriage return characters. -Listfromline = Line.split ('\ t')#split Line into an element list -Returnmat[index,:] = Listfromline[0:3]#Select the first 3 elements to store in the feature matrix +Classlabelvector.append (int (listfromline[-1]))#Select the last element to store -Index + = 1 +     returnReturnmat, Classlabelvector A  at if __name__=='__main__': -Datingdatamat, datinglabels = File2matrix ('DatingTestSet2.txt') -Fig =plt.figure () -Ax = Fig.add_subplot (111) -Ax.scatter (datingdatamat[:,0],datingdatamat[:,1], 20.0*array (datinglabels), 15.0*Array (datinglabels)) -Plt.show ()

Import NumPy as NP>>> Np.zeros ((3,2)) #3行2列的零矩阵Array ([[0.,  0.],       [0.,  0.],       [0.,  0.]])

Ax.scatter (datingdatamat[:,0],datingdatamat[:,1], 20.0*array (datinglabels), 15.0*array (DatingLabels))#  Scatter (x, y, size, color)

" "normalized eigenvalue newvalue = (oldvalue-min)/(max-min)" "defAutonorm (dataSet): Minvals=dataset.min (0) maxvals=Dataset.max (0) ranges= Maxvals-minvals Normdataset=zeros (Shape (dataSet)) m=Dataset.shape[0] Normdataset= Dataset-tile (Minvals, (M, 1))#Use the tile function to copy the contents of the variable into the same size as the input matrixNormdataset = Normdataset/tile (ranges, (M, 1))    returnnormdataset, ranges, minvalsif __name__=='__main__': Datingdatamat, Datinglabels= File2matrix ('DatingTestSet2.txt') Normmat, ranges, minvals=autonorm (datingdatamat) FIG=plt.figure () Ax= Fig.add_subplot (111) Ax.scatter (normmat[:,0],normmat[:,1], 30.0*array (datinglabels), 100.0*Array (datinglabels)) Plt.xlabel (U'number of frequent flyer miles earned per year') Plt.ylabel (U'percentage of time spent playing video games') plt.show ()

1 " "2 test the code. 3 estimate the correct rate of the algorithm: provide 90% of the existing sample as a training sample, and use the remaining 10% data to test the classifier. 4 Error Rate = Number of error results/Total number of test data5 " "6 defdatingclasstest ():7HoRatio = 0.108Datingdatamat, datinglabels = File2matrix ('DatingTestSet2.txt')9Normmat, ranges, minvals =autonorm (Datingdatamat)Tenm =Normmat.shape[0] Onenumtestvecs = Int (M * hoRatio)#10% of the sample count is used for testing AErrorcount =0 -      forIinchRange (numtestvecs): -Classifierresult =classify0 (Normmat[i,:], normmat[numtestvecs:m,:], theDATINGLABELS[NUMTESTVECS:M], 20) -         Print "The classifier came back with:%d, the real answer is:%d"  -%(Classifierresult, datinglabels[i]) -         if(Classifierresult! = Datinglabels[i]): Errorcount + = 1.0 +     Print "The total error rate is:%f"% (Errorcount/float (numtestvecs)) #输出错误率 -  + if __name__=='__main__': ADatingclasstest ()

1 " "2 Enter a person's information and give a forecast of their preferred level3 " "4 defClassifyperson ():5Resultlist = [' not at all','In small doses','In large doses']6Percenttats = Float (raw_input ("percentage of time spend playing video games?"))7Ffmiles = Float (Raw_input("frequent flier miles earned per year?"))8Icecream = Float (raw_input ("liters of ice cream consumed per year?"))9Datingdatamat, datinglabels = File2matrix ('DatingTestSet2.txt')TenNormmat, ranges, minvals =autonorm (Datingdatamat) OneInarr =Array ([Ffmiles, Percenttats, icecream]) AClassifierresult = Classify0 ((inarr-minvals)/ranges, Normmat, Datinglabels, 3) -     Print "You'll probably like this person :", resultlist[classifierresult-1] -  the if __name__=='__main__': -Classifyperson ()

Test:

1 percentage of time spend playing video games?102 frequent flier miles earned per year?100003 Liters of ice cream consumed per year?0.54 You'll probably like this person: in   small doses

Machine learning Combat--the second chapter of improving the pairing effect of dating sites