osdir.com


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

pandas loc on str lower for column comparison


Sayth Renshaw <flebber.crue at gmail.com> writes:

> On Sunday, 1 September 2019 10:48:54 UTC+10, Sayth Renshaw  wrote:
>> I've created a share doc same structure anon data from my google drive.
>> 
>> https://drive.google.com/file/d/0B28JfFTPNr_lckxQRnFTRF9UTEFYRUVqRWxCNVd1VEZhcVNr/view?usp=sharing
>> 
>> Sayth
>
> I tried creating the df1 dataframe by using iloc instead of loc to avoid any column naming issues.
>
> So i created a list of integers for iloc representing the columns in current example.
>
> df1 = df.iloc[[0,1,5,6,7]] 
>
> However, I ust be misunderstanding the docs
> https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.iloc.html#pandas.DataFrame.iloc
> Allowed inputs are:
>
> An integer, e.g. 5.
> A list or array of integers, e.g. [4, 3, 0].
>
> Because while it works I appear to grab all columns 13 when I requested 5.
> 	UID Name FTE Agent ID Current Leader New Leader Current Team New
> Team Current Site New Site Unnamed: 10 Unnamed: 11 Unnamed: 12
>
> How do I misunderstand iloc?
>
That would select ROWS 0,1,5,6,7, not columns.
To select columns 0,1,5,6,7, use two-dimensional indexes

df1 = df.iloc[:, [0,1,5,6,7]]

: selects all rows.
-- 
Piet van Oostrum <piet-l at vanoostrum.org>
WWW: http://piet.vanoostrum.org/
PGP key: [8DAE142BE17999C4]