0

I took this example to run interactive Bokeh (in this case with sliders) in Jupyter Notebook: How to get interactive bokeh in Jupyter notebook

That works fine.

But when I start the notebook with voila on the command line ("voila bokeh_jupyter.ipynb") than the only output I see is "BokehJS 2.4.2 successfully loaded."

Why does it not work with Voila and what can I do about it?

Marc8932
  • 115
  • 10
  • You'd have to provide some more information to get much help. Your link is to a question with code that didn't quite work. So which code did you use as a basis for yours? The not quite-working-code or [version at the bottom here](https://stackoverflow.com/a/59585360/8508004)? And can you share a toy version via mybinder? You also probably want to see @krassowski 's below [here](https://stackoverflow.com/q/70497685/8508004), the one dated Dec 29 '21 at 5:17. – Wayne Jan 27 '22 at 17:22
  • yes, it's the exact version at the bottom, under "and the modified version of your code:". It runs fine in the Jupyter Notebook, but I can not run it with Voila. The question is actualy part of a larger context/question, which is: I want to be able to run nice-looking interactive dashboard (with sliders, checkboxes, linking plots, etc..) in Jupyter Notebook, without bothering the enduser with code. Based on packages in the Python ecosystem that are future-proof / require little maintanance and codelines. – Marc8932 Jan 28 '22 at 09:32
  • What I stumbled upon are (combination of) packages like: Plotly, Bokeh, Streamlit, Datapane, Panel, iPywidgets, Holoviz, Voila. I'm looking for something that fits in the requirements above and is broadly supported / future proof. I think Voila/Jupyter/Bokeh would be a good candidate. But until so far I can't get it running with Voila. But maybe there is a better solution.. – Marc8932 Jan 28 '22 at 09:42
  • If it has all the features you seek, Voila and Jupyter are better integrated together. For a demo of sliders controlling a plot in Voila, go [here](https://github.com/fomightez/3Dscatter_plot-binder), press `launch binder`. When the session comes you'd want to step through running the cells ( launching in the case of the last one) the last three demonstrations listed on the page that comes up. That may give you a better idea of combining widgets and voila and plots. Or move towards using the app dashboard of Bokeh or Holoviz panel. – Wayne Jan 28 '22 at 14:30

0 Answers0