如何在 Spark SQL 中为每个组创建 z-score [英] How to create a z-score in Spark SQL for each group

问题描述

我有一个看起来像这样的数据框

I have a dataframe which looks like this

        dSc     TranAmount
 1: 100021      79.64
 2: 100021      79.64
 3: 100021       0.16
 4: 100022      11.65
 5: 100022       0.36
 6: 100022       0.47
 7: 100025       0.17
 8: 100037       0.27
 9: 100056       0.27
10: 100063       0.13
11: 100079       0.13
12: 100091       0.15
13: 100101       0.22
14: 100108       0.14
15: 100109       0.04

现在我想用每个 TranAmount 的 z-score 创建第三列，这将是

Now I want to create a third column with the z-score of each TranAmount which will be

(TranAmount-mean(TranAmount))/StdDev(TranAmount)

这里的平均值和标准偏差将基于每个 dSc 的组

here mean and standard deviation will be based on groups of each dSc

现在我可以在 Spark SQL 中计算均值和标准差了.

Now I can calculate mean and standard deviation in Spark SQL.

(datafromdb
  .groupBy("dSc")
  .agg(datafromdb.dSc, func.avg("TranAmount") ,func.stddev_pop("TranAmount")))

但我不知道如何使用数据框中的 z 分数实现第三列.我将不胜感激任何指向实现这一目标的正确方法/

but I am at a loss on how to achieve a third column with the z-score in the data frame. I would appreciate any pointer to the right way of achieving this/

推荐答案

例如，您可以计算统计数据并join 原始数据:

You can for example compute statistics and join with the original data:

stats = (df.groupBy("dsc")
  .agg(
      func.stddev_pop("TranAmount").alias("sd"), 
      func.avg("TranAmount").alias("avg")))

df.join(broadcast(stats), ["dsc"])

(df
    .join(func.broadcast(stats), ["dsc"])
    .select("dsc", "TranAmount", (df.TranAmount - stats.avg) / stats.sd))

或使用窗口函数带有标准偏差公式:

from pyspark.sql.window import Window
import sys

def z_score_w(col, w):
    avg_ = func.avg(col).over(w)
    avg_sq = func.avg(col * col).over(w)
    sd_ = func.sqrt(avg_sq - avg_ * avg_)
    return (col - avg_) / sd_

w = Window().partitionBy("dsc").rowsBetween(-sys.maxsize, sys.maxsize)
df.withColumn("zscore", z_score_w(df.TranAmount, w))

这篇关于如何在 Spark SQL 中为每个组创建 z-score的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！