In pyspark, after reading data from impala/hive, how is it partitioned?

Question

I would like to make sure my assumption is right about partitions right after reading data.

By default, a RDD in my cluster has 200 partitions.

I read data fro