那些遇到过的问题

使用 pandas_udf 时出现“索引值为空”错误

niu*_*uer 5 apache-spark-sql pyspark

对于 pyspark 中的 DataFrame，如果使用 F.lit(1) （或任何其他值）初始化列，则它会被分配给 pandas_udf 内的某些值（在本例中使用 shift()，但也可能发生在任何其他函数上） ），这会导致“索引处的值为空”错误。

谁能提供一些提示为什么会发生这种情况？这是 pyspark 中的错误吗？

请参阅下面的代码和错误。

spark = SparkSession.builder.appName('test').getOrCreate()
df = spark.createDataFrame([Row(id=1, name='a', c=3),
Row(id=2, name='b', c=6),
Row(id=3, name='a', c=2),
Row(id=4, name='b', c=9),
Row(id=5, name='c', c=7)])

df = df.withColumn('f', F.lit(1))


@pandas_udf(df.schema, PandasUDFType.GROUPED_MAP)
def shift_test(pdf):
    pdf['f'] = pdf['c'].shift(1)
    return pdf

df = df.groupby(['name']).apply(shift_test)
df.show()
Run Code Online (Sandbox Code Playgroud)

f如果我将列设置为等于c 请参阅下面的输出，则不会出现此类错误。

+---+---+----+---+
|  c| id|name|  f|
+---+---+----+---+
|  3|  1|   a|  1|
|  6|  2|   b|  1|
|  2|  3|   a|  1|
|  9|  4|   b|  1|
|  7|  5|   c|  1|
+---+---+----+---+


---------------------------------------------------------------------------
Py4JJavaError                             Traceback (most recent call last)
<ipython-input-46-5b4a8c6e0258> in <module>
     18 
     19 df = df.groupby(['name']).apply(shift_test)
---> 20 df.show()

Py4JJavaError: An error occurred while calling o3378.showString.
: org.apache.spark.SparkException: Job aborted due to stage failure: Task 97 in stage 426.0 failed 4 times, most recent failure: Lost task 97.3 in stage 426.0 (TID 6258, optoldevny1, executor 0): java.lang.IllegalStateException: Value at index is null
    at org.apache.arrow.vector.IntVector.get(IntVector.java:101)
    at org.apache.spark.sql.vectorized.ArrowColumnVector$IntAccessor.getInt(ArrowColumnVector.java:299)
    at org.apache.spark.sql.vectorized.ArrowColumnVector.getInt(ArrowColumnVector.java:84)
    at org.apache.spark.sql.execution.vectorized.MutableColumnarRow.getInt(MutableColumnarRow.java:117)
    at org.apache.spark.sql.catalyst.expressions.GeneratedClass$SpecificUnsafeProjection.apply(Unknown Source)
    at org.apache.spark.sql.catalyst.expressions.GeneratedClass$SpecificUnsafeProjection.apply(Unknown Source)
    at scala.collection.Iterator$$anon$11.next(Iterator.scala:410)
    at org.apache.spark.sql.catalyst.expressions.GeneratedClass$GeneratedIteratorForCodegenStage3.processNext(Unknown Source)
    at org.apache.spark.sql.execution.BufferedRowIterator.hasNext(BufferedRowIterator.java:43)
    at org.apache.spark.sql.execution.WholeStageCodegenExec$$anonfun$13$$anon$1.hasNext(WholeStageCodegenExec.scala:636)
    at org.apache.spark.sql.execution.SparkPlan$$anonfun$2.apply(SparkPlan.scala:255)
    at org.apache.spark.sql.execution.SparkPlan$$anonfun$2.apply(SparkPlan.scala:247)
    at org.apache.spark.rdd.RDD$$anonfun$mapPartitionsInternal$1$$anonfun$apply$24.apply(RDD.scala:858)
    at org.apache.spark.rdd.RDD$$anonfun$mapPartitionsInternal$1$$anonfun$apply$24.apply(RDD.scala:858)
    at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
    at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:346)
    at org.apache.spark.rdd.RDD.iterator(RDD.scala:310)
    at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
    at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:346)
    at org.apache.spark.rdd.RDD.iterator(RDD.scala:310)
    at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:90)
    at org.apache.spark.scheduler.Task.run(Task.scala:123)
    at org.apache.spark.executor.Executor$TaskRunner$$anonfun$10.apply(Executor.scala:408)
    at org.apache.spark.util.Utils$.tryWithSafeFinally(Utils.scala:1360)
    at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:414)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
    at java.lang.Thread.run(Thread.java:748)

Driver stacktrace:
    at org.apache.spark.scheduler.DAGScheduler.org$apache$spark$scheduler$DAGScheduler$$failJobAndIndependentStages(DAGScheduler.scala:1891)
    at org.apache.spark.scheduler.DAGScheduler$$anonfun$abortStage$1.apply(DAGScheduler.scala:1879)
    at org.apache.spark.scheduler.DAGScheduler$$anonfun$abortStage$1.apply(DAGScheduler.scala:1878)
    at scala.collection.mutable.ResizableArray$class.foreach(ResizableArray.scala:59)
    at scala.collection.mutable.ArrayBuffer.foreach(ArrayBuffer.scala:48)
    at org.apache.spark.scheduler.DAGScheduler.abortStage(DAGScheduler.scala:1878)
    at org.apache.spark.scheduler.DAGScheduler$$anonfun$handleTaskSetFailed$1.apply(DAGScheduler.scala:927)
    at org.apache.spark.scheduler.DAGScheduler$$anonfun$handleTaskSetFailed$1.apply(DAGScheduler.scala:927)
    at scala.Option.foreach(Option.scala:257)
    at org.apache.spark.scheduler.DAGScheduler.handleTaskSetFailed(DAGScheduler.scala:927)
    at org.apache.spark.scheduler.DAGSchedulerEventProcessLoop.doOnReceive(DAGScheduler.scala:2112)
    at org.apache.spark.scheduler.DAGSchedulerEventProcessLoop.onReceive(DAGScheduler.scala:2061)
    at org.apache.spark.scheduler.DAGSchedulerEventProcessLoop.onReceive(DAGScheduler.scala:2050)
    at org.apache.spark.util.EventLoop$$anon$1.run(EventLoop.scala:49)
    at org.apache.spark.scheduler.DAGScheduler.runJob(DAGScheduler.scala:738)
    at org.apache.spark.SparkContext.runJob(SparkContext.scala:2061)
    at org.apache.spark.SparkContext.runJob(SparkContext.scala:2082)
    at org.apache.spark.SparkContext.runJob(SparkContext.scala:2101)
    at org.apache.spark.sql.execution.SparkPlan.executeTake(SparkPlan.scala:365)
    at org.apache.spark.sql.execution.CollectLimitExec.executeCollect(limit.scala:38)
    at org.apache.spark.sql.Dataset.org$apache$spark$sql$Dataset$$collectFromPlan(Dataset.scala:3389)
    at org.apache.spark.sql.Dataset$$anonfun$head$1.apply(Dataset.scala:2550)
    at org.apache.spark.sql.Dataset$$anonfun$head$1.apply(Dataset.scala:2550)
    at org.apache.spark.sql.Dataset$$anonfun$52.apply(Dataset.scala:3370)
    at org.apache.spark.sql.execution.SQLExecution$$anonfun$withNewExecutionId$1.apply(SQLExecution.scala:80)
    at org.apache.spark.sql.execution.SQLExecution$.withSQLConfPropagated(SQLExecution.scala:127)
    at org.apache.spark.sql.execution.SQLExecution$.withNewExecutionId(SQLExecution.scala:75)
    at org.apache.spark.sql.Dataset.withAction(Dataset.scala:3369)
    at org.apache.spark.sql.Dataset.head(Dataset.scala:2550)
    at org.apache.spark.sql.Dataset.take(Dataset.scala:2764)
    at org.apache.spark.sql.Dataset.getRows(Dataset.scala:254)
    at org.apache.spark.sql.Dataset.showString(Dataset.scala:291)
    at sun.reflect.GeneratedMethodAccessor67.invoke(Unknown Source)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:498)
    at py4j.reflection.MethodInvoker.invoke(MethodInvoker.java:244)
    at py4j.reflection.ReflectionEngine.invoke(ReflectionEngine.java:357)
    at py4j.Gateway.invoke(Gateway.java:282)
    at py4j.commands.AbstractCommand.invokeMethod(AbstractCommand.java:132)
    at py4j.commands.CallCommand.execute(CallCommand.java:79)
    at py4j.GatewayConnection.run(GatewayConnection.java:238)
    at java.lang.Thread.run(Thread.java:748)
Caused by: java.lang.IllegalStateException: Value at index is null
    at org.apache.arrow.vector.IntVector.get(IntVector.java:101)
    at org.apache.spark.sql.vectorized.ArrowColumnVector$IntAccessor.getInt(ArrowColumnVector.java:299)
    at org.apache.spark.sql.vectorized.ArrowColumnVector.getInt(ArrowColumnVector.java:84)
    at org.apache.spark.sql.execution.vectorized.MutableColumnarRow.getInt(MutableColumnarRow.java:117)
    at org.apache.spark.sql.catalyst.expressions.GeneratedClass$SpecificUnsafeProjection.apply(Unknown Source)
    at org.apache.spark.sql.catalyst.expressions.GeneratedClass$SpecificUnsafeProjection.apply(Unknown Source)
    at scala.collection.Iterator$$anon$11.next(Iterator.scala:410)
    at org.apache.spark.sql.catalyst.expressions.GeneratedClass$GeneratedIteratorForCodegenStage3.processNext(Unknown Source)
    at org.apache.spark.sql.execution.BufferedRowIterator.hasNext(BufferedRowIterator.java:43)
    at org.apache.spark.sql.execution.WholeStageCodegenExec$$anonfun$13$$anon$1.hasNext(WholeStageCodegenExec.scala:636)
    at org.apache.spark.sql.execution.SparkPlan$$anonfun$2.apply(SparkPlan.scala:255)
    at org.apache.spark.sql.execution.SparkPlan$$anonfun$2.apply(SparkPlan.scala:247)
    at org.apache.spark.rdd.RDD$$anonfun$mapPartitionsInternal$1$$anonfun$apply$24.apply(RDD.scala:858)
    at org.apache.spark.rdd.RDD$$anonfun$mapPartitionsInternal$1$$anonfun$apply$24.apply(RDD.scala:858)
    at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
    at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:346)
    at org.apache.spark.rdd.RDD.iterator(RDD.scala:310)
    at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
    at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:346)
    at org.apache.spark.rdd.RDD.iterator(RDD.scala:310)
    at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:90)
    at org.apache.spark.scheduler.Task.run(Task.scala:123)
    at org.apache.spark.executor.Executor$TaskRunner$$anonfun$10.apply(Executor.scala:408)
    at org.apache.spark.util.Utils$.tryWithSafeFinally(Utils.scala:1360)
    at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:414)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
    ... 1 more
Run Code Online (Sandbox Code Playgroud)

Str*_*keR 5

您收到错误的原因是由于函数null引入的值shift，其次是因为您没有对返回的架构进行任何更改以接受这些null值。

当您将返回模式指定为 时，spark 默认情况下会从原始模式中df.schema获取。nullable = False因此，您需要在此处为​​ column 提供一个新架构f，您需要在其中进行设置nullable = True以避免此错误。

# Schema of output DataFrame
new_schema = StructType([
    StructField("c", IntegerType(), False), 
    StructField("id", IntegerType(), False), 
    StructField("name", StringType(), False), 
    StructField("f", IntegerType(), True)
  ])

@pandas_udf(new_schema, PandasUDFType.GROUPED_MAP)
def shift_test(pdf):
    pdf['f'] = pdf['c'].shift(1)
    return pdf
Run Code Online (Sandbox Code Playgroud)

niu*_*uer 2

看起来 pyspark 当来自 pandas_udf 时无法处理缺失值。在进入 pandas_udf 之前，它期望每列具有某种数据类型（如 @pandas_udf(df.schema, PandasUDFType.GROUPED_MAP).

如果存在任何缺失值（可以在shift此处生成），则会抛出异常，因为 Java 无法处理缺失值（例外是 Java 例外：）java.lang.IllegalStateException。

要解决此问题，需要将此类缺失值替换为正确类型的值，此处为integer.

这个新pandas_udf功能解决了这个问题：

@pandas_udf(df.schema, PandasUDFType.GROUPED_MAP)
def shift_test(pdf):
    pdf['f'] = pdf['c'].shift(1)
    pdf['f'].fillna(value=-1, inplace=True) #replace missing values with -1
    return pdf
Run Code Online (Sandbox Code Playgroud)

这是输出

+---+---+----+---+
|  c| id|name|  f|
+---+---+----+---+
|  7|  5|   c| -1|
|  6|  2|   b| -1|
|  9|  4|   b|  6|
|  2|  3|   a| -1|
|  3|  1|   a|  2|
+---+---+----+---+
Run Code Online (Sandbox Code Playgroud)

归档时间：

查看次数：

3272 次

最近记录：

4 年，5 月 前
相关归档
Spark Dataframe验证镶木地板写入(scala)的列名称 7
SPARK - 为什么 Spark 作业持续时间不等于每个阶段持续时间的总和？ 6
整个数据框上的 Pyspark 窗口函数 5
withColumn不允许我使用max()函数生成新列 4
如何根据数据类型识别列并在pyspark中进行转换？ 4
KeyError:SparkConf初始化期间的SPARK_HOME 3
如何使用 pyspark 启动独立集群？ 3
在编写spark UDF(而不是将UDF作为一个)时会有性能损失吗？ 3
如何过滤pyspark数据帧中的空值？ 3
在spark-SQL而不是spark-scala中创建“临时”表？ 0
难疑归档
如何从JavaScript中删除数组中的特定元素？ 7655
如何将新的本地分支推送到远程Git存储库并跟踪它？ 4154
在Git存储库中查找并恢复已删除的文件 2716
如何通过引用传递变量？ 2480
如何在Java中调用另一个构造函数？ 2144
将字符串转换为datetime 2035
如何在Ruby中编写switch语句 2026
如何设置HTML <select>元素的默认值？ 1343
用于更新和删除的HTTP状态代码？ 1264
如何检查对象是否在JavaScript中有密钥？ 1047