Two versions of Python

[testb...@gmail.com]

Hi,

I'm using both python 2 and 3 in my JupyterHub and I have two separate kernels (json file)  for each one of them.

Could anyone please clarify on the below?

1. The Python 3 is displayed  as "Python [default]" in my JupyterHub in the list of notebooks.

Where do I have to modify to display it as "Python 3.5" in my JupyterHub in the list of notebooks.

2. When I install a python package, what command can I use to get it installed in both python 2 (anaconda2) and python3 (anaconda3)?

Do I need to run conda install command in ananconda 2 and anaconda3?

Thanks in advance!

[Thomas Kluyver]

Hi,

On 7 August 2017 at 11:16, <testb...@gmail.com> wrote:

1. The Python 3 is displayed  as "Python [default]" in my JupyterHub in the list of notebooks.

Where do I have to modify to display it as "Python 3.5" in my JupyterHub in the list of notebooks.

The "[default]" bit probably means you're using the nb_conda_kernels extension, which replaces our handling of kernelspecs with a mechanism to find kernels in conda envs. That can be convenient, but if you'd rather control the available kernels with kernel.json files, you can remove the package.
 

2. When I install a python package, what command can I use to get it installed in both python 2 (anaconda2) and python3 (anaconda3)?

Do I need to run conda install command in ananconda 2 and anaconda3?

Yup. Each Python has its own separate collection of installed packages, and installing a package adds it to whichever one you're using to install it.