Pysyft: Inline Documentation Consistency

[iamtrask]

User Story: As a regular Contributor to OpenMined's PySyft repository, I want documentation to follow a consistent format to ease my use of the platform. Presently, this is not the case.

Acceptance Criteria:
• a defined template for inline documentation
• modifying PySyft repositories to conform to this template (particularly math.py and tensor.py which are highly irregular)

[CT83]

Assuming Documentation like that of functions equal(TensorBase,TensorBase) & masked_select() in tensor.py is expected. I am on it !

[drewvolpe]

@CT83 I would recommend we use an existing style for docstrings. The two most popular ones are Google style and Numpy style. Both are well-supported by Sphinx and tools like PyLint and Jedi. Given we're using Numpy in various parts of the code, I would vote for NumPy style.

http://sphinxcontrib-napoleon.readthedocs.io/en/latest/example_numpy.html

[iamtrask]

@CT83 very excited about this!!!!

[CT83]

I don't know if you guys know, but you were featured here! https://youtu.be/HAC6sqq7_-U?t=7m13s
I was interested, excited and wanted to help.

[CT83]

I don't really know how this works but I worked on one of the two files and I am done;
https://github.com/OpenMined/PySyft/pull/273 Let me know if what I did was, what was expected and I will do the math.py File too. If not I am open to Advice :)

[momonala]

Hey! I'm new to Openmined and this seems like an issue I can tackle. To those working on the issue - how can I help out?

[CT83]

@iamtrask @drewvolpe https://github.com/OpenMined/PySyft/pull/273 I have completed tensor.py should I start with math.py? in the same way?

[bharathgs]

@CT83 tensor.py looks great, I think you can start off with math.py.