How to specify local kernel on remote repository?

[risaueno]

Hi!

I'm in a remote respostory using Remote-SSH, and I'm trying to run my .py files with #%% cell tags using the interactive window, but using an already running kernel e.g. localhost:8888.

The interactive window allows me to select a remote conda environment to create a kernel to run this (on the top right). But I can only see options for remote conda environments. Is it possible to use a kernel running on my local machine localhost:8888:?token=xxx, which I can access through my local browser?

Using port forwarding I have set it up so that the kernel I want to connect to is available from port 1111 on this remote machine. I tried "Jupyter: Specify jupyter server for connections" -> specify the URI of an existing server" and typing in "https://localhost:1111/?token=xxx" but this just hangs and does nothing.

I suppose this is the Atom editor Hydrogen equivalent where I could just specify the jupyter address & token which the editor looks for from my local machine and connect to it. Is there such functionality in VScode?

[risaueno]

Update:
Because the kernel was made available locally on this remote machine, this did eventually work: "Jupyter: Specify jupyter server for connections" -> specify the URI of an existing server" and typing in "http://localhost:1111/?token=xxx", I had to remove "https" and replace it with "http". It seemed to do nothing and I couldn't select it from the already opened interactive window, but when I closed it and ran the the first cell, it opened a new interactive window with the correct kernel (with label Python 3 (ipykernel)).

However I would still like to know if it is possible to connect to an actual local kernel i.e. kernel running on my machine, instead of this "localhost" from the perspective of the remote machine. But perhaps this is a limitation related to Remote-SSH.