2020年10月22日

This how I currently keep myself updated in the ever growing world of Python.

To manage multiple python versions I use pyenv:

  pyenv install --list
  pyenv install VERSION

(e.g. pyenv install 3.9.0)

Then, to change the global python version use:

  pyenv global VERSION

(e.g. pyenv global 3.9.0)

These are the libraries I normally use at the present time, and I use pip to manage them:

  pip install numpy matplotlib pandas
              opencv-python jupyterlab scipy
              scikit-learn scikit-image seaborn

Note for myself:

If installing python for the first time, remember to add .pyenv/bin directory to $PATH and pyenv init to the shell configuration as well.

Checking compatibility with Jupyter

Now, in order to use the latest version of python’s kernel in jupyter, I have to modify alfre/Library/Jupyter/kernels/python3/kernel.json, specifying the version of python that I’m actually using.

To check which version of python jupyter is actually using, open a console inside jupyter’s environment and type:

  import sys
  print(sys.version)

Uninstalling old versions

  pyenv uninstall VERSION

(e.g. pyenv uninstall 3.8.5)