I also have this question. I know that you can integrate R into tableau, but it would be great if there was a way to trigger the execution of a script in say, scikit-learn, and pass the resulting data frame back.
Here's a good overview of how Tableau can integrate with R -> Tableau and R Integration
As for Python, it's not something that we currently support. It is something that we are investigating since there are more & more requests for it.
1 of 1 people found this helpful
Hi Bora Beran & ALL,
I have a urgent similar request,
- Creating parameter and user gonna provide input
- Passing the input to python script (~100 lines to generate table with multiple columns ~80)
- automatically execute python script after receiving the user input
- Display the results as crosstab below the parameters in the same dashboard
Please let me know whether is this is possible.
I have an idea
- From parameter filter, we can pass the value to python using SCRIPT_STR(tabpy.query),
- By using extract API for executing those 100 lines of python scripts
I think, once we pass the value to python, it can execute the scripts automatically based on that and generate extract API, i.e..hyperfile,
But the question is whether we can make the hyper in a way like a "Live connection" and display it in the same dashboard where we are displaying the filters?
Let me know whether this is feasible or is there is any other way to make it happen?
Thanks in Advance,
Hi Renga, have you made this working? Could you share it with the community?
Wonder if you made this happen, and could you share the solution with the community?
Jiayu Lin & @David Shi,
But found that we can create .tde files using python,
So in our python script, we need to add the script to generate .tde files (which i believe is changed to .hyper now)
And use the entire python code as calculated field, and pass the required values as parameters, so that user can provide their values,
and based on that this calculated field(python code) will execute and generate the .tde file.
But here is where we got stuck, so we need to replace this new .tde file with the old file in the server.
And at the same time, the extract refresh schedule is 15 mins and the user need to wait for at least 15 mins to see the results,
So this approach is dropped.