Interactive Forecasting in Power BI: User Input, Python Model, and Dynamic Visualization

[Anonymous]

Hello all,

I'm seeking guidance on implementing an interactive forecasting feature in Power BI Service (browser version). Here's what I'm trying to achieve:

1. User Input: Allow business users to input a value for a specific field directly in the Power BI Service interface.
2. Python Integration: Use this input as a variable in a custom Python forecasting model.
3. Dynamic Visualization: Display the forecasting results back in the Power BI report, updating visualizations based on the user's input.

Specific Requirements:

• Time series data forecasting using a custom Python model
• Users should be able to select a field (corresponding to a model feature) and input a proposed value for a specific date
• The Python model should use this input to generate predictions
• Predictions should be plotted alongside actual data in Power BI

Key Questions:

1. Is this possible to implement entirely within Power BI?
2. What's the most efficient way to handle user input in Power BI Service?
3. How can I integrate Python execution with user inputs in real-time?
4. What's the best method to update visualizations dynamically based on Python output?

I'm open to exploring Power Apps, Power Automate, or any other Microsoft Power Platform tools if they can help achieve this functionality.

I would greatly appreciate any insights, suggestions, or alternative approaches. Please let me know if you need any additional information about my project.

Thank you in advance for your help!

[lbendlin]

1. Heck no

2. There are a couple of shady (at best) third party visuals that allow you to collect user input and write back into the data source (via their own servers, shudder)

3. Python visuals can react to dataframe changes, but be reminded that they need to render something at the end of the processing

4. not sure what you mean by that - please elaborate.