Is it possible to use jupyter-ipywidgets+ipympl as a Matplotlib backend?

[vadimcn]

Often, I need to visualize data from the debug console while debugging Python scripts on a remote server (via VSCode's SSH remote).
While it is possible to use X11 forwarding + Matplotlib's interactive backends, this combination is often so slow as to be completely unusable.
So I am wondering: what would it take to use jupyter-ipywidgets+ipympl as a Matplotlib backend from a regular Python script?

[DonJayamanne]

jupyter-ipywidgets+ipympl as a Matplotlib backend from a regular Python script?

ipympl is supported today and should work,
Please could you let me know what problems you are running into when attempting to use that

[vadimcn]

How do I use it?
Please note that I'd like to do ad-hoc plotting form the VSCode debug console while debugging a regular Python app, not a notebook. Basically, something similar to what's being demoed here.

[DonJayamanne]

How do I use it?

• Create a ntoebook
• Add some code in the cell with ipympl and matplot lib
• Then run the code and you will get the output

lease note that I'd like to do ad-hoc plotting form the VSCode debug console w

Currently its not possible to open ipywidgets outputs outside a notebook view.
You are after this feature #1120
Please upvote that issue.

[vadimcn]

#1120 seems to be about moving plot from a notebook out into a separate panel? But I would like to plot from outside of a notebook view, so it doesn't seem to solve my problem...