Updated intro documentation covering delta methods.

[daryadedik]

Updated .ipynb and generated new .html file.

[gbordyugov]

Looks great, but that's exactly the problem with this kind of documentation: it changes a lot (styles, formatting, etc.) after small changes.

In the future we have to think of a better alternative.

[coveralls]

Coverage remained the same at 82.828% when pulling 077640c17f736d5dff009f4346a3cb97ded6d47d on daryaded_expan into 66e1aa3eeee477e42adfc03daaa7143da96c64d4 on dev.

[daryadedik]

AFAIK, in R there is an automated documentation file generation process, you don't need to do formatting you just change fields (name of function, what it does, what parameters it accepts) and create examples and build it (very similar to latex format). Don't know whether there is something like that in Python.

[shansfolder]

That's the documentation generated by Sphinx we had in the /docs folder. It can read doc strings and build an api doc, but we are not using it. We should definitely plan for a ticket to have better documentations (not the quick fix we have right now).

[shansfolder]

and for the current PR, I have two quick comments:

1. Can you move the section for parameters of deltas up, before the section of interpreting results?
2. Regarding the fields in metadata, I actually meant removing only baseline_variant (and keep the others as optional, maybe only experiment as mandatory) because in the example we have more:

   {'experiment': 'random_data_generation',
   'primary_KPI': 'normal_shifted',
   'source': 'simulated'}

[daryadedik]

@shansfolder sure, will do the changes.

[coveralls]

Coverage remained the same at 82.828% when pulling 6dd1d1c1369b784b1e7733b386d6bf03d68af73c on daryaded_expan into 66e1aa3eeee477e42adfc03daaa7143da96c64d4 on dev.