More a question than an issue actually. I'm currently working on the rppg branch of this package and I followed the latest "standard" we agreed upon for docstrings (i.e. numpy style https://sphinxcontrib-napoleon.readthedocs.io/en/latest/example_numpy.html).
Problem is that most of (if not all) existing code's docstrings does not follow this format. As a consequence, it looks like my numpystyle docstrings are not parsed properly by sphinx.
So my question is: should we stick to the original docstring format for the package, or should we move everything to numpystyle docstrings ?