41

我想知道是否有某种方法可以为多列上的 spark 数据帧指定自定义聚合函数。

我有一个像这样的类型（名称，项目，价格）的表：

john | tomato | 1.99
john | carrot | 0.45
bill | apple  | 0.99
john | banana | 1.29
bill | taco   | 2.59

至：

我想将项目和每个人的成本汇总到这样的列表中：

john | (tomato, 1.99), (carrot, 0.45), (banana, 1.29)
bill | (apple, 0.99), (taco, 2.59)

这在数据框中可能吗？我最近了解到，collect_list但它似乎只适用于一栏。

4

5 回答 5

98

考虑struct在收集为列表之前使用该函数将列组合在一起：

import org.apache.spark.sql.functions.{collect_list, struct}
import sqlContext.implicits._

val df = Seq(
  ("john", "tomato", 1.99),
  ("john", "carrot", 0.45),
  ("bill", "apple", 0.99),
  ("john", "banana", 1.29),
  ("bill", "taco", 2.59)
).toDF("name", "food", "price")

df.groupBy($"name")
  .agg(collect_list(struct($"food", $"price")).as("foods"))
  .show(false)

输出：

+----+---------------------------------------------+
|name|foods                                        |
+----+---------------------------------------------+
|john|[[tomato,1.99], [carrot,0.45], [banana,1.29]]|
|bill|[[apple,0.99], [taco,2.59]]                  |
+----+---------------------------------------------+
于 2017-03-10T19:50:46.370 回答
35

作为 a 执行此操作的最简单方法DataFrame是先收集两个列表，然后UDF将zip两个列表一起使用。就像是：

import org.apache.spark.sql.functions.{collect_list, udf}
import sqlContext.implicits._

val zipper = udf[Seq[(String, Double)], Seq[String], Seq[Double]](_.zip(_))

val df = Seq(
  ("john", "tomato", 1.99),
  ("john", "carrot", 0.45),
  ("bill", "apple", 0.99),
  ("john", "banana", 1.29),
  ("bill", "taco", 2.59)
).toDF("name", "food", "price")

val df2 = df.groupBy("name").agg(
  collect_list(col("food")) as "food",
  collect_list(col("price")) as "price" 
).withColumn("food", zipper(col("food"), col("price"))).drop("price")

df2.show(false)
# +----+---------------------------------------------+
# |name|food                                         |
# +----+---------------------------------------------+
# |john|[[tomato,1.99], [carrot,0.45], [banana,1.29]]|
# |bill|[[apple,0.99], [taco,2.59]]                  |
# +----+---------------------------------------------+
于 2016-06-10T11:37:52.180 回答
10

也许比zip函数更好的方法（因为 UDF 和 UDAF 对性能非常不利）是将两列包装到Struct.

这可能也会起作用：

df.select('name, struct('food, 'price).as("tuple"))
  .groupBy('name)
  .agg(collect_list('tuple).as("tuples"))
于 2018-08-06T17:38:58.690 回答
4

就您而言，collect_list 似乎仅适用于一列：要使 collect_list 在多列上工作，您必须将您想要作为聚合的列包装在结构中。例如：

     val aggregatedData = df.groupBy("name").agg(collect_list(struct("item", "price")) as("food"))

     aggregatedData.show
+----+------------------------------------------------+
|name|foods                                           |
+----+------------------------------------------------+
|john|[[tomato, 1.99], [carrot, 0.45], [banana, 1.29]]|
|bill|[[apple, 0.99], [taco, 2.59]]                   |
+----+------------------------------------------------+
于 2020-03-02T04:14:28.310 回答
2

这是一个选项，将数据框转换为 Map 的 RDD，然后groupByKey在其上调用 a。结果将是一个键值对列表，其中 value 是一个元组列表。

df.show
+----+------+----+
|  _1|    _2|  _3|
+----+------+----+
|john|tomato|1.99|
|john|carrot|0.45|
|bill| apple|0.99|
|john|banana|1.29|
|bill|  taco|2.59|
+----+------+----+


val tuples = df.map(row => row(0) -> (row(1), row(2)))
tuples: org.apache.spark.rdd.RDD[(Any, (Any, Any))] = MapPartitionsRDD[102] at map at <console>:43

tuples.groupByKey().map{ case(x, y) => (x, y.toList) }.collect
res76: Array[(Any, List[(Any, Any)])] = Array((bill,List((apple,0.99), (taco,2.59))), (john,List((tomato,1.99), (carrot,0.45), (banana,1.29))))
于 2016-06-10T02:20:48.440 回答