Algebird Hyper Log Log

hyperLogLogExample.scala

import com.twitter.algebird.HyperLogLogMonoid

//define test data
val data = Seq("aaa", "bbb", "ccc")
//create algebird HLL
val hll = new HyperLogLogMonoid(bits = 10)
//convert data elements to a seq of hlls
val hlls = data.map { str =>
  val bytes = str.getBytes("utf-8")
  hll.create(bytes)
}
//or:
val data = List(1, 1, 2, 2, 3, 3, 4, 4, 5, 5)
val hlls = data.map { hllMonoid.create(_) }

//WARN: don`t use merged.size - it is a different thing
//get the estimate count from merged hll
println("estimate count: " + hll.sizeOf(merged).estimate)
//or
println("estimate count: " + merged.approximateSize.estimate)