Cloudera Data Science Workbench supports using Spark 2 from Python via PySpark.
pi.py
What kind of feedback do you have?
