file.write(str(matrix[i][featNo])+" ")
file.write("\n")
return 0
previousRecord = [4,5]
hotelProp = [8,9,10,11,12,13,15,16]
typeOftrip = [18,19,20,21,22,23]
etc = [24,25]
comp = range(27,51)
relativeFreq = [54,55,56,57]
oneHotHotel = range(58,110)
sumComp = [110,111,112]
ratio = range(113,117)
allData = hh.getCSVmatrix("smalltrainOld&New2",100000)
trainData = allData[:3*(len(allData))/4]
evalData = allData[3*(len(allData))/4::]
print len(trainData)
print len(evalData)
trainData = hh.convertMatrixtoFloat(trainData)
evalData = hh.convertMatrixtoFloat(evalData)
writeForSVMRank('train',trainData)
writeForSVMRank('test',evalData)
step1Output = "smalltestProcessedStep1"
step2Output = "smalltestRelFreq"
step3_1Output = "smalltestProcessedStep3"
step3_2Output = "smalltestRatios"
step4_FinalOutput = "smalltestOld&New2"
step1 = 0
step2 = 0
step3 = 0
step4 = 0
step5 = 0
step6 = 0
if (step1 == 1):
print "Step1 - Fill Null"
allData = hh.getCSVmatrix(firstInput,totalRows)
searchIDs = hh.getColumn(allData,0)
#Date_time - Get weeks is [[0,..0,1,0],[0,..1,0,0]... ,[1,0,..]]
date = hh.getColumn(allData,1)
booking = np.array(hh.getColumn(allData,19))
date = hh.getColumn(allData,1)
for i in range(len(date)):
time = date[i]
time = (time.split())[0]
month = int(time[5:7])
day = int(time[-11:-9])
dictMonth = {1:31,2:59,3:90,4:120,5:151,6:181,7:212,8:243,9:273,10:304,11:334,12:365}
num_week = (dictMonth[month] + day+ int(booking[i]) )
if num_week>=365:
num_week = num_week%365
num_week = num_week//7
