spark之QuantileDiscretizer源码解析

1、前言：

这里重点讲一下transform函数。

2、正文

直接上源码：

/**
   * Binary searching in several buckets to place each data point.
   * @param splits array of split points
   * @param feature data point
   * @param keepInvalid NaN flag.
   *                    Set "true" to make an extra bucket for NaN values;
   *                    Set "false" to report an error for NaN values
   * @return bucket for each data point
   * @throws SparkException if a feature is < splits.head or > splits.last
   */

  private[feature] def binarySearchForBuckets(
      splits: Array[Double],
      feature: Double,
      keepInvalid: Boolean): Double = {
    if (feature.isNaN) {
      if (keepInvalid) {
        splits.length - 1
      } else {
        throw new SparkException("Bucketizer encountered NaN value. To handle or skip NaNs," +
          " try setting Bucketizer.handleInvalid.