Cowa

class Car {
  constructor(make, wheels) { 
    this.make = make;
    this.wheels = wheels;
    this.currentSpeed = 25;
  }

  printCurrentSpeed() {
    console.log(this.make + ' is going ' + this.currentSpeed + ' mph.');
  }

Cowa

// Tokenized text
val words = List("bonjour", "je", "suis", "ici", ".", "bonjour", "je", "suis", "là")

// 2-grams
val n = 2
val ngram = words.sliding(n).toList
// ngram: List[List[String]] = List(List(bonjour, je), List(je, suis), List(suis, ici), List(ici, .), List(., bonjour), List(bonjour, je), List(je, suis), List(suis, là))

val ngramWithCount = ngram.groupBy(identity).mapValues(_.size)
// ngramWithCount: Map[List[String],Int] = Map(List(bonjour, je) -> 2, List(je, suis) -> 2, List(suis, ici) -> 1, List(suis, là) -> 1, List(ici, .) -> 1, List(., bonjour) -> 1)

Cowa

val source = "The blue sky is near the red koala near the blue sky"
val tokens = source.split(" ").toList
// List(The, blue, sky, is, near, the, red, koala, near, the, blue, sky)

Cowa

// For bigrams
tokens.sliding(2).toList
// List(List(The, blue), List(blue, sky), List(sky, is), List(is, near), List(near, the), List(the, red), List(red, koala), List(koala, near), List(near, the), List(the, blue), List(blue, sky))

// For trigrams
tokens.sliding(3).toList
// List(List(The, blue, sky), List(blue, sky, is), List(sky, is, near), List(is, near, the), List(near, the, red), List(the, red, koala), List(red, koala, near), List(koala, near, the), List(near, the, blue), List(the, blue, sky))

// For n-gram
val n = ...

Cowa

val n = 2
val ngram = tokens.sliding(n).toList
// List(List(The, blue), List(blue, sky), List(sky, is), List(is, near), List(near, the), List(the, red), List(red, koala), List(koala, near), List(near, the), List(the, blue), List(blue, sky))

Cowa

val ngramWithCount = ngram.groupBy(identity).mapValues(_.size)
// Map(List(koala, near) -> 1, List(blue, sky) -> 2, List(red, koala) -> 1, List(near, the) -> 2, List(the, blue) -> 1, List(The, blue) -> 1, List(sky, is) -> 1, List(the, red) -> 1, List(is, near) -> 1)

Cowa

val ngramWithProbability = ngramWithCount.map { case (k, v) =>
  (k, v.toDouble / ngramWithCount.filterKeys(_.take(n - 1) == k.take(n - 1)).values.sum)
}
// Map(List(eating, the) -> 1.0, List(blue, sky) -> 1.0, List(red, koala) -> 1.0, List(near, the) -> 1.0, List(The, blue) -> 1.0, List(the, sky) -> 0.5, List(koala, eating) -> 1.0, List(sky, is) -> 1.0, List(the, red) -> 0.5, List(is, near) -> 1.0)

Cowa

val sumIndex = ngramWithCount.groupBy { case (k, v) => k.take(n - 1) }.mapValues(_.values.sum)

val ngramWithProbabilityFaster = ngramWithCount.map { case (k, v) => 
  (k, v.toDouble / sumIndex(k.take(n - 1)))
}
// Map(List(eating, the) -> 1.0, List(blue, sky) -> 1.0, List(red, koala) -> 1.0, List(near, the) -> 1.0, List(The, blue) -> 1.0, List(the, sky) -> 0.5, List(koala, eating) -> 1.0, List(sky, is) -> 1.0, List(the, red) -> 0.5, List(is, near) -> 1.0)

Cowa

val ngram = (List.fill(n - 1)("<s>") ++ tokens :+ "</s>")
  .sliding(n)
  .toList

// Same song
val ngramWithCount = ...
val ngramWithProbabilityFaster = ...

Cowa

object Marker {
  val start = "<s>"
  val end = "</s>"
}

object NGram {
  def build(t: String, n: Int): NGram = {
    NGram(addProbability(addCount(t.split("\n").flatMap(l =>
      sliding(l.split(" ").toList, n)).toList), n), n)
  }