val inputRDD = sc.textFile(path = ".../fakefriends.csv")

    def parseLineIntoTuple(line: String): (Int, Int) = {
      val parts = line.split(",")
      val age = parts(2).toInt
      val numFriends = parts(3).toInt
      (age, numFriends)
    }

    val avgFriendsByAge = inputRDD
      .map(parseLineIntoTuple)                                  // (age, friends)
      .mapValues(v => (v, 1))                                   // (age, (friends, 1))
      .reduceByKey((v1, v2) => (v1._1 + v2._1, v1._2 + v2._2))  // (age, (f1+f2+..., 1+1...) sum freinds, count freq
      .mapValues(v => v._1 / v._2)                              // (age, sum / count)
      .sortBy(t => t._1)                                        // by first thing in tuple (age)
      .collect()                                                // RDD -> Array

    avgFriendsByAge.foreach(println(_))