Thanks, Matthias. That's helpful.

As a quick followup, I need to do:
python - m ipykernel install --name=<envname_or_similar>

Within each conda environment that I want available, right?
-paul

On Thu, Apr 6, 2017 at 2:53 PM, Matthias Bussonnier <
[email protected]> wrote:

> Hi Paul.
>
> Up until recently, conda was shipping with nb_kernel_extension (or I
> don't remember the name) that would automatically discover kernel in
> multiple env. So 2 is IMHO (prefered).
>
> Also at some point the notebook _server_ will drop Python 2, so you
> will have to run the server from a separate env from your Python 2
> kernels. That basically whal all the non-python kernel are doing.
>
> So 2 seem to be the prefered way if the only thing you care about are
> kernels extensions.
>
> Nuance on your "2", the notebook does not need to be in the root env,
> it can be on any env, and see kernels from other envs.
> --
> M
>
> On Thu, Apr 6, 2017 at 1:54 PM, Paul Hobson <[email protected]> wrote:
> > Jupyter team:
> >
> > Congrats on the 5.0 release. The new notebook is sooooo nice. All of the
> > hard work is very much appreciated.
> >
> > I'm curious what folks (users and devs alike) have settled on as their
> best
> > practices for managing jupyter and ipykernels when you have many conda
> > environments.
> >
> > At the moment I see two general approaches:
> > 1) install everything into every conda environment where they are needed,
> > and launch the notebook server from that environment
> >
> > or
> >
> > 2) only install jupyter and notebook into the root (default) conda
> > environment, install ipykernel into the remaining environments, and
> launch
> > jupyter from the default env, selecting the kernel as necessary.
> >
> > Is either one of those a particularly bad idea? Is there a different
> > approach I missed?
> >
> > -Paul
> >
> > --
> > You received this message because you are subscribed to the Google Groups
> > "Project Jupyter" group.
> > To unsubscribe from this group and stop receiving emails from it, send an
> > email to [email protected].
> > To post to this group, send email to [email protected].
> > To view this discussion on the web visit
> > https://groups.google.com/d/msgid/jupyter/CADT3MEDT4Qx7sqFo6CnHrwzcTwS1M
> BSYtnouVT5ku-BqSW9-kw%40mail.gmail.com.
> > For more options, visit https://groups.google.com/d/optout.
>
> --
> You received this message because you are subscribed to the Google Groups
> "Project Jupyter" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> To post to this group, send email to [email protected].
> To view this discussion on the web visit https://groups.google.com/d/
> msgid/jupyter/CANJQusVHC05Y5nZmNdfjDh0HAy71C%3D6kYtg%2B69AgOZ7rJpVLGg%
> 40mail.gmail.com.
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups 
"Project Jupyter" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/jupyter/CADT3MEC%2BEVBpTFYczyZFwe8aAvy3sspR%2BLny_wF0YrrTzuwLDg%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to