Outlier Elimination in Spark With InterQuartileRange Results in Error-CodePudding

I have the following recursive function that determines the Outlier using the InterQuartileRange method:

def interQuartileRangeFiltering(df: DataFrame): DataFrame = {
    @scala.annotation.tailrec
    def inner(cols: List[String], acc: DataFrame): DataFrame = cols match {
      case Nil          => acc
      case column :: xs =>
        val quantiles = acc.stat.approxQuantile(column, Array(0.25, 0.75), 0.0) // TODO: values should come from config
        println(s"$column ${quantiles.size}")
        val q1 = quantiles(0)
        val q3 = quantiles(1)
        val iqr = q1 - q3
        val lowerRange = q1 - 1.5 * iqr
        val upperRange = q3   1.5 * iqr
        val filtered = acc.filter(s"$column < $lowerRange or $column > $upperRange")
        inner(xs, filtered)
    }
    inner(df.columns.toList, df)
}


val outlierDF = interQuartileRangeFiltering(incomingDF)

So basically what I'm doing is that I'm recursively iterating over the columns and eliminating the outliers. Strangely it results in an ArrayIndexOutOfBounds Exception and prints the following:

housing_median_age 2
inland 2
island 2
population 2
total_bedrooms 2
near_bay 2
near_ocean 2
median_house_value 0
java.lang.ArrayIndexOutOfBoundsException: 0
  at inner$1(<console>:75)
  at interQuartileRangeFiltering(<console>:83)
  ... 54 elided

What is wrong with my approach?

CodePudding user response：

Here is what I came up with and works like a charm:

def outlierEliminator(df: DataFrame, colsToIgnore: List[String])(fn: (String, DataFrame) => (Double, Double)): DataFrame = {

    val ID_COL_NAME = "id"
    val dfWithId = DataFrameUtils.addColumnIndex(spark, df, ID_COL_NAME)
    val dfWithIgnoredCols = dfWithId.drop(colsToIgnore: _*)

    @tailrec
    def inner(
      cols: List[String],
      filterIdSeq: List[Long],
      dfWithId: DataFrame
    ): List[Long] = cols match {
      case Nil          => filterIdSeq
      case column :: xs =>
        if (column == ID_COL_NAME) {
          inner(xs, filterIdSeq, dfWithId)
        } else {
          val (lowerBound, upperBound) = fn(column, dfWithId)
          val filteredIds =
            dfWithId
              .filter(s"$column < $lowerBound or $column > $upperBound")
              .select(col(ID_COL_NAME))
              .map(r => r.getLong(0))
              .collect
              .toList
          inner(xs, filteredIds    filterIdSeq, dfWithId)
        }
    }

    val filteredIds = inner(dfWithIgnoredCols.columns.toList, List.empty[Long], dfWithIgnoredCols)
    dfWithId.except(dfWithId.filter($"$ID_COL_NAME".isin(filteredIds: _*)))
  }