Join us at FabCon Atlanta from March 16 - 20, 2026, for the ultimate Fabric, Power BI, AI and SQL community-led event. Save $200 with code FABCOMM.
Register now!The Power BI Data Visualization World Championships is back! Get ahead of the game and start preparing now! Learn more
Hello there
Am a newbie to preditive analytics world.
I wonder now that we can run R script as part of ETL within Power BI. How does this give pros and cons if say I want to run general linear regression or random forest using R script comparing to azure ml -anyone has had compared the accuracy of results in both approaches?
Please can you share your views?
Thank you 🙂
Hi @Anonymous,
Do you have more insights on using R vs Azure ML for predictive analytics? Did you try either, was wondering if there is anything to learn from your experience?
@cyclist007 hi there sorry I just saw your message.
Yes I have explored a lot between Azure ML and R. I dont claim to be an expert but below are just my views
1. Azure ML - if you have end users who will consume the model you build through powerbi or excel then Azure ML is a great tool. The pre defined modules (e.g transformation, normalise, ML algorithms and so one) are easy to use.
2. R - I use PowerBI at work on day to day basis and along side that Im learning R so I try to find any opportunity to use both.... I think PowerBI is almost like a Microsoft answer to Shiny - but the downside of using Microsoft PowerBI for R visualisation is that not many packages are supported by PowerBI. Package like ggplot2 is great but I would like to see Microsoft PowerBi takes on html widget and stuff....
So i sometimes use PowerBI in ETL process then use R to do modelling and stuff.
I dont visualise R within PowerBI as we need Pro license for that and we are on budget 😛
As per model accuracy / precision and so on - I found them to be very similar. So i think its not the tool that judges whether my models are good or not its how I feature engineer my variable and tune my model....so that this point Azure ML or R are immaterial in a way....
So in short - for a quick prototype - I use PowerBI + R, once I trust my models to be stable and consumable I move on to Azure ML then publish this as web service then consume it via PowerBI.
Hope this helps? Share me your experience I am keen to know too. Many thanks
@Anonymous Using power bi desktop via query editor -> Transform tab -> Run R Scripts, you can run R scripts to perform tasks you want via R. Note that this will create dataset for you that you can use to execute R scripts on.
Hi Pedzilla,
More information on Running Machine Learning through Azure ML in PowerBI can be found here.
Thanks & Regards,
Bhavesh
The Power BI Data Visualization World Championships is back! Get ahead of the game and start preparing now!
| User | Count |
|---|---|
| 38 | |
| 36 | |
| 33 | |
| 33 | |
| 29 |
| User | Count |
|---|---|
| 134 | |
| 96 | |
| 78 | |
| 67 | |
| 65 |