You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Michael H (JIRA)" <ji...@apache.org> on 2017/11/28 16:07:00 UTC
[jira] [Comment Edited] (SPARK-22629) incorrect handling of calls
to random in UDFs
[ https://issues.apache.org/jira/browse/SPARK-22629?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16268960#comment-16268960 ]
Michael H edited comment on SPARK-22629 at 11/28/17 4:06 PM:
-------------------------------------------------------------
Hi Sean,
Thanks for looking into this! I was worried this would be by design...
Just to confirm then, if running the below:
{code:none}
df_br = spark.createDataFrame([{'name': 'hello'}])
df_br.withColumn('RAND', udf_random_col()).withColumn('RAND_PLUS_TEN', udf_add_ten('RAND')).show()
+-----+----+-------------+
| name|RAND|RAND_PLUS_TEN|
+-----+----+-------------+
|hello| 31| 74|
+-----+----+-------------+
{code}
is it expected that the second withColumn call happens via a full re-evaluation?
Many thanks!
was (Author: micheusch):
Hi Sean,
Thanks for looking into this! I was worried this would be by design...
Just to confirm then, if running the below:
{code:none}
df_br = spark.createDataFrame([{'name': 'hello'}])
df_br.withColumn('RAND', udf_random_col()).withColumn('RAND_PLUS_TEN', udf_add_ten('RAND')).show()
{code}
is it expected that the second withColumn call happens via a full re-evaluation?
Many thanks!
> incorrect handling of calls to random in UDFs
> ---------------------------------------------
>
> Key: SPARK-22629
> URL: https://issues.apache.org/jira/browse/SPARK-22629
> Project: Spark
> Issue Type: Bug
> Components: PySpark
> Affects Versions: 2.1.0
> Reporter: Michael H
>
> {code:none}
> df_br = spark.createDataFrame([{'name': 'hello'}])
> # udf creates a random integer
> udf_random_col = udf(lambda: int(100*random.random()), IntegerType())
> # add a column to our DF using that udf
> df_br = df_br.withColumn('RAND', udf_random_col())
> df_br.show()
> +-----+----+
> | name|RAND|
> +-----+----+
> |hello| 68|
> +-----+----+
> # udf that adds 10 to an input column value
> random.seed(1234)
> udf_add_ten = udf(lambda rand: rand + 10, IntegerType())
> # unexpected result due to re-evaluation
> df_br.withColumn('RAND_PLUS_TEN', udf_add_ten('RAND')).show()
> +-----+----+-------------+
> | name|RAND|RAND_PLUS_TEN|
> +-----+----+-------------+
> |hello| 72| 87|
> +-----+----+-------------+
> # workaround: cache the resulst after using the random number generating udf
> df_br.withColumn('RAND', udf_random_col()).cache().withColumn('RAND_PLUS_TEN', udf_add_ten('RAND')).show()
> +-----+----+-------------+
> | name|RAND|RAND_PLUS_TEN|
> +-----+----+-------------+
> |hello| 68| 78|
> +-----+----+-------------+
> {code}
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org