Python virtual enviroment

Python virtual environments (venv) are good to:

1. Use a specific version of Python;
2. Do pip install rightly (without sudo);
3. Be able to use other Python versions as jupyter notebook run kernel.

It is the recommended way of organizing packages versions in Python 3.3+.
To create/activate/deactivate a venv:

python3.8 -m venv ~/python3.8-venv
source ~/python3.8-venv/bin/activate
deactivate

In an activated venv, install pip and then do pip installs, e.g.:

python3.8 -m pip install --user --upgrade pip
pip install opencv-python jupyter notebook

(It's a good package to have "import cv2" without compiling OpenCV.)

To install jupyter kernels, do:

ipython kernel install --name "python3.8-venv" --user