Following along in Scala with the Coursera Machine Learning course

MachineLearning1.scala

import spire.implicits._
import spire.math._

class MachineLearning1[Number](implicit fractional: Fractional[Number]) {

  import fractional.pow

  private def number(integer: Int): Number = fractional.fromInt(integer)

  type Hypothesis = Number => Number
  type Cost = Hypothesis => Number
  type TrainingSet = Set[(Number, Number)]

  def hypothesis(constant: Number)(weightX: Number): Hypothesis = {
    x => constant + weightX * x
  }

  def cost(trainingSet: TrainingSet)(hypothesis: Hypothesis): Number = {
    val averager = number(1) / (number(2) * number(trainingSet.size))
    val totalError = trainingSet.foldLeft(number(0)) { case (error, (input, actual)) =>
      val prediction = hypothesis(input)
      error + pow(prediction - actual, 2)
    }
    val averageError = averager * totalError
    averageError
  }

  def gradientDescent(learningRate: Number)(cost: Cost): Number = {
    // Generate hypotheses and feed them to cost



  }

  def minimize(cost: Cost): Number = {
    ???
  }

}