I am running a pyspark application from EMR cluster. I have two buckets -an input data bucket user-data-input and an o/p data bucket to store the script results