How to change jupyter kernel from Python 2 to python 3?

You can do this with the following steps:

1. conda create -n py36 'python=3.6' ipykernel #Replace 3.6 with desired version.

2. To activate installed jupyter kernal you need run, source activate py36

3. python -m ipykernel install --user

4. The interesting part: if you want to switch between kernels (py2-py3) in the same notebook, you need to run, conda install nb_conda

However, if at any point you realize that some of the modules are not available, for that you need to check Anaconda Python version.

python - version

if it is not python 3.x, you need to run

conda create -n py36 python=3.6 anaconda
source active py36

I hope it helps and enjoy switching python versions in the same notebook. You can try to print something in both python2 and python 3.

You can do this with the following steps:

1. Create new env by using python3.

   virtualenv -p /usr/bin/python3.7 jupyter_environment

2. Activate env

   source jupyter_environment/bin/activate

3. Start jupyter

   jupyter notebook --allow-root

Then open Jupyter notebook, you will find python3 by default on your kernel.

I have succeeded to add Python 3.9 to Jupyter as follows.

# cd /usr/share/jupyter/kernels
# ls
python3
# cp python3 python3.9
# cd python3.9

and changed python3 in the kernel.json file to python3.9:

# cat kernel.json 
{
 "argv": [
  "/usr/bin/python3.9",
  "-m",
  "ipykernel_launcher",
  "-f",
  "{connection_file}"
 ],
 "display_name": "Python 3.9",
 "language": "python"

After this, the kernel may be changed through Kernel->Change kernel in Jupyter.

Python 3.9 was installed separately, simply with package manager (zypper in my case).

I think, this should work with Python 2 as well.

In case you installed the python/jupyter with anaconda, simply update the python from 2.* to 3.*