Spark directory partition name

[Mohit Anchlia]

When spark writes the partition it writes in the format as:

<partition key name1>=<keyvalue/<partition key name 2>/key value>

Is there a way to have spark write only by keyvalue?