Complete research of Pandas dataframe model and API

[rufuspollock]

Have already been doing research as partially summarized in the README. Need to complete this ...

• key APIs
• differences between Pandas and R

[s-celles]

xarray http://xarray.pydata.org/ (previously named xray) is probably also an API and an implementation to consider

Pandas Panel will probably be removed in favor of xarray.DataArray
see github.com/pandas-dev/pandas/issues/8906

Some R datasets such as

• Titanic
• HairEyeColor

are available at https://github.com/Rdatasets/python

[s-celles]

Some questions about how to handle data in JS can be found here

https://www.quora.com/Is-there-a-data-handling-library-like-Python-Panda-for-JavaScript
http://stackoverflow.com/questions/30610675/python-pandas-equivalent-in-javascript
http://stackoverflow.com/questions/23451828/data-analysis-with-javascript/23451897#23451897

See also
http://learnjsdata.com/
Pinging @vlandham @grssnbchr @ndarville @bijanbwb @iros

[s-celles]

You might also have a look at https://json-stat.org/ and https://github.com/badosa/JSON-stat
Pinging @badosa

[rufuspollock]

@scls19fr thanks - really useful.

[s-celles]

Some sources using such a format are given here datasets/awesome-data#181 (comment)

[badosa]

@rgrp The API of the JSON-stat Javascript Toolkit (JJT) can be found here https://github.com/badosa/JSON-stat/wiki/API-Reference This API allows you to handle cubes of data (rather than data) in the JSON-stat format.

pyjstat https://pypi.python.org/pypi/pyjstat/ allows you handle JSON-stat documents in python using pandas DataFrames. For R you have https://github.com/ajschumacher/rjstat

The JSON-stat Javascript Utilities Suite (JJUS) allows you to convert to https://github.com/badosa/JSON-stat/blob/master/utils/tocsv.md and from https://github.com/badosa/JSON-stat/blob/master/utils/fromcsv.md CSV (but probably not in the kind of structure you would expect as, again, JSON-stat is focused on cubes, not data).