AttributeError: 'NoneType' Object Has No Attribute 'setCallSite'

In PySpark, I want to calculate the correlation between two dataframe vectors, using the following code (I do not have any problem in importing pyspark or createDataFrame): from py

Solution 1:

There's an open resolved issue around this:

https://issues.apache.org/jira/browse/SPARK-27335?jql=text%20~%20%22setcallsite%22

[Note: as it's resolved, if you're using a more recent version of Spark than October 2019, please report to Apache Jira if you're still encountering this issue]

The poster suggests forcing to sync your DF's backend with your Spark context:

df.sql_ctx.sparkSession._jsparkSession = spark._jsparkSession
df._sc = spark._sc

This worked for us, hopefully can work in other cases as well.

Solution 2:

I got the same error not only with Correlation.corr(...) dataframe, but with ldaModel.describeTopics() as well.

Most probably it is the SPARK bug.

They forget to initialise DataFrame::_sc._jsc member when created resulting dataframe.

Each dataframe has normally this member initialised with proper JavaObject.

Solution 3:

There are several reasons for getting that AttributeError:

1. You can use sc.stop before initializing one of xContext (where x could be SQL, Hive). For example:

   sc = SparkContext.getOrCreate(conf = conf)  
   sc.stop() 
   spark = SQLContext(sc)  
   

2. Your spark is not synchronized on a cluster.

So, just restart your jupyter notebook kernel or reboot an application (not spark context) and it will work.