au.csiro.variantspark.utils.FactorVariable.scala Maven / Gradle / Ivy

Go to download

Show more of this group Show more artifacts with this name
Show all versions of variant-spark_2.11 Show documentation

Genomic variants interpretation toolkit

The newest version!

package au.csiro.variantspark.utils

case class FactorVariable(values: Array[Int], nCategories: Int,
    subset: Option[Array[Int]] = None) {

  def indices: Iterator[Int] =
    subset.map(_.toIterator).getOrElse(values.indices.toIterator)

  def counts: Array[Int] = {
    val labelCounts = Array.fill(nCategories)(0)
    indices.foreach(i => labelCounts(values(i)) += 1)
    labelCounts
  }

  def apply(indexes: Array[Int]): FactorVariable = {
    new FactorVariable(values, nCategories, Some(indexes))
  }
}

object FactorVariable {
  def apply(values: Array[Int]): FactorVariable = apply(values, values.max + 1)

  def classCounts(currentSet: Array[Int], labels: Array[Int], labelCount: Int): Array[Int] = {
    val labelCounts = Array.fill(labelCount)(0)
    currentSet.foreach(i => labelCounts(labels(i)) += 1)
    labelCounts
  }
  def labelMode(currentSet: Array[Int], labels: Array[Int], labelCount: Int): Int = {
    val labelCounts = classCounts(currentSet, labels, labelCount)
    ArraysUtils.maxIndex(labelCounts)
  }
}