Issue with python loc on dataframe

Question

I'm currently working on a project on python with jupyter notebook. I would like to predict stadium attendance in france (ligue 1).

To achieve that I have taken data from web with beautiful soup. I'm trying now to clean up my data : I have some missing values for the stadiums and I would like to assign stadium for a specific team (Olympique lyonnais).

I first tried that :

stats_match.stade[(stats_match.saison >= 2017) & (stats_match.domicile == 'Olympique Lyonnais') & (stats_match.stade.isna())] = 'Groupama stadium'

which gave me that error :

SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame

So I followed the instruction and tried that :

stats_match.stade.loc((stats_match.saison >= 2017) & (stats_match.domicile == 'Olympique Lyonnais') & (stats_match.stade.isna())) = 'Groupama stadium'

which give me :

File "", line 3 stats_match.stade = stats_match.stade.loc((stats_match.saison >= 2017) & (stats_match.domicile == 'Olympique Lyonnais') & (stats_match.domicile.isna())) = 'Groupama stadium'
SyntaxError: can't assign to function call

What do I miss here ? Do I have to use .where function ? Many thanks

The first issue is that .loc uses square brackets, not .loc() — roganjosh
– roganjosh, Commented Mar 1, 2019 at 15:27
Second, I don't think you want to be calling this on the series, but rather on the df itself — roganjosh
– roganjosh, Commented Mar 1, 2019 at 15:29
I do, because If I call it on the dataframe It would assign my value on all the columns, no ? — Tiboo
– Tiboo, Commented Mar 1, 2019 at 15:37
No, because loc allows you to specify the column that you want to transform, separated from your conditions by a comma but still inside the square brackets — roganjosh
– roganjosh, Commented Mar 1, 2019 at 15:38
It works thanks ! didn't know I could specify the colunm in the loc function. — Tiboo
– Tiboo, Commented Mar 1, 2019 at 15:41

Rajat Jain · Accepted Answer · 2019-03-01 15:35:39Z

1

This should be correct;

stats_match.loc[(stats_match.saison >= 2017) & (stats_match.domicile == 'Olympique Lyonnais') & (stats_match.stade.isna())] = 'Groupama stadium'

answered Mar 1, 2019 at 15:35

Rajat Jain

2,0422 gold badges17 silver badges23 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Tiboo · Accepted Answer · 2019-03-01 15:34:16Z

0

Okay like @roganjosh said I put () instead of [], but I have now the same warning "SettingWithCopyWarning, I already had this warning and used .copy(), but is there a better way of doing it ?

answered Mar 1, 2019 at 15:34

Tiboo

274 bronze badges

Collectives™ on Stack Overflow

Issue with python loc on dataframe

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related