Using Python with Power BI for Enterprise-wide Reports

[cwollett]

I am looking for feedback from others who publish reports for an organization.

We have had a few projects crop up where using Python and/or R would be a huge benefit:

• There's no available visualization built-in or via the app search that does what we want but we could build one in Python or R that does.
• We want to request data from an API and transform it in a way that would be easier for us via Python than trying to do it via Power Query

Power BI Desktop has all the tools needed to do these things with Python, but you cannot refresh a Power BI report that uses Python/R scripts.

How are others handling that?

• Are you just not using Python?
• Are you using Data Factory?
• Are you using a "personal gateway" in place of an enterprise gateway? Is there a performance hit if you do this?

[lbendlin]

The requirement for a personal gateway immediately disqualifies this for enterprise use. 

There are two more aspects you need to consider.

Power BI is a one way street. It has no solid technical foundation for writing data back in to other systems.

When you use Python or R visuals you are restricted to the default renderer.  You cannot bring your own.  Python visuals are not interactive, R visuals can be made interactive with some contortions.  Neither of them is capable of contributing to the filter context.

You can use Python and R in your enterprise, just not together with Power BI.