I am trying to use accumulators in pyspark to instrument my udf\'s or custom spark methods in my pyspark jobs. I have written a custom SparkListener in java/scala to listen to t
