Forum Discussion
Errow when scheduling refresh for Python script in Power BI
Hi Polly,
I downloaded the latest Anaconda distribution a few days ago so I suppose I am using Python 3.9.
I have created a new env called powerbi_env and have added the path to the new env as shown below. I added the pandas, matplotlib, numpy packages through the Anaconda navigator (also shown below). It could be run using Power BI desktop without issues. However, when I try to use the personal gateway to automate it, an error pops up as shown in the below first screencap. It seems the gateway cannot connect to the most new environment? Any ideas what I should do?
Hi Anonymous ,
Please refer to the topics to see if them help.
Python script source error- ADO.net failed to load
- returned the Gateway version to 3000.37.35.
Power BI python script env change worked for PowerBI Desktop but not work for app.powerbi.com
- adding new environment directory as path variable.
Best Regards
Community Support Team _ Polly
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
- Anonymous4 years agoNot applicable
Hi Polly,
I try the first method and download the oldest version I can find in this page 3000.101.16. But it still pops up the same error message.
For adding new environment directory as path variable. Could you please describe what I should exactly do in the process?
- Anonymous4 years agoNot applicable
Hi Anonymous ,
Step by step procedure:
- Download last version of Winpython from : https://winpython.github.io/
- Install Winpython on C: \
- In Power BI Desktop, select File > Options and settings > Options > Python scripting. The Python script options page appears.
- Specify the local Python installation path with the Winpython folder defined at step 2
- Restart Power BI
More details: Power BI Python with Anaconda missing dependency
Best Regards
Community Support Team _ Polly
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.