def parseLine(line):
fields = line.split(',')
age = int(fields[2])
numFriends = int(fields[3])
return (age,numFriends)

line = sparkCont.textFile("D:\ResearchInMotion\ml-100k\fakefriends.csv")
rdd = line.map(parseLine)
totalsByAge = rdd.mapValues(lambda x: (x, 1)).reduceByKey(lambda x, y: (x[0] + y[0], x[1] + y[1]))
averagesByAge = totalsByAge.mapValues(lambda x: x[0] / x[1])
results = averagesByAge.collect()
for result in results:
    print(result)