Hi all! I would like to ask for your help to understand how to work with personal Python packages in the JupyterLab workspace. I created a conda environment and installed some of my packages. However, the environment does not appear in JupyterLab's kernel list. Is it possible to access personal environments from a JupyterLab session? Thanks!
Dylan Nelson
13 Jul
The Lab environment is just like a home directory on any linux computer. You can install anything you wish.
You can certainly install, and/or use, conda, and you can create/use conda environments. You just need to make sure that everything you do is inside your home directory, by setting appropriate paths, etc. Probably you can google for some details of this in conda.
To actually get such a new environment to be listed in the Lab kernel list would be possible I think. JupyterLab is also just an installed python package (inside your Lab), so you can change its configuration. If you change it very much, or if you change the version of the package, things will break and the Lab will no longer work. But I think by following a tutorial on how to add a kernel, you should be able to make it work.
However, I'm not sure it's worth it. Perhaps easier is to just activate your conda environment on the command line and use it there?
Hi all! I would like to ask for your help to understand how to work with personal Python packages in the JupyterLab workspace. I created a conda environment and installed some of my packages. However, the environment does not appear in JupyterLab's kernel list. Is it possible to access personal environments from a JupyterLab session? Thanks!
The Lab environment is just like a home directory on any linux computer. You can install anything you wish.
You can certainly install, and/or use, conda, and you can create/use conda environments. You just need to make sure that everything you do is inside your home directory, by setting appropriate paths, etc. Probably you can google for some details of this in conda.
To actually get such a new environment to be listed in the Lab kernel list would be possible I think. JupyterLab is also just an installed python package (inside your Lab), so you can change its configuration. If you change it very much, or if you change the version of the package, things will break and the Lab will no longer work. But I think by following a tutorial on how to add a kernel, you should be able to make it work.
However, I'm not sure it's worth it. Perhaps easier is to just activate your conda environment on the command line and use it there?