DataFrame/Series.describe()

[rxin]

pandas version: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.describe.html

This is a very useful function for understanding the dataset. Note that Spark already has a describe that does similar things. A simple version of Koalas' describe can just delegate to Spark's. A fancier version should construct the descriptive stats and then match more pandas'.

[tdliu]

Is someone looking into this? We have a webinar coming up and it would be great to be able to use this method (cc @HyukjinKwon, @ueshin).

[rxin]

Yes please. We should use approximate quantile in Spark automatically.

…

On Tue, May 21, 2019 at 5:52 PM tdliu ***@***.***> wrote: Is someone looking into this? We have a webinar coming up and it would be great to be able to use this method (cc @HyukjinKwon <https://github.com/HyukjinKwon>, @ueshin <https://github.com/ueshin>). — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#334?email_source=notifications&email_token=AACO6PA64NGMLUWIQ2QHM43PWQZHJA5CNFSM4HM7XIEKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODV4VY5I#issuecomment-494492789>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AACO6PC3ZVBRBFZ5LNYLDZLPWQZHJANCNFSM4HM7XIEA> .

[garawalid]

@tdliu I'll try to finish it ASAP.