CODEWITHSUNDEEP
pythonpandasloops
SLV
SLV

Reputation: 23

I can't figure out how to loop with this error : a value is trying to be set on a copy of a slice from a DataFrame

Hi :) I would like to know if for each modality (df.A == 'First' and df.B = 'Second'), if for the column df.B the rows are equal (or not) to the first row (df['B'][0]).

A B C
First 1 None
First 2 None
First 3 None
Second 2 None
Second 2 None
Second 2 None

The output should be :

A B C
First 1 None
First 2 False
First 3 False
Second 2 None
Second 2 True
Second 2 True

I've tried this code :

def equal(df) :
    arr = df['A'].unique()
    arr = arr.tolist()
    dfd = df.copy()
    for i in arr : 
        for m in range(1, len(dfd['A']==i)) :
            if dfd['B'][m] == dfd['B'][0] :
                dfd['C'][m] == True
            else :
                dfd['C'][m] == False
equal(HoldingsLegal)

But I got this error :

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

Thank you for your help.

Upvotes: 1

Views: 48

Answers (1)

Pablo C
Pablo C

Reputation: 4761

IIUC this is what you want:

df['C'] = df.groupby('A')['B'].transform("nunique").eq(1)
#        A  B      C
#0   First  1  False
#1   First  2  False
#2   First  3  False
#3  Second  2   True
#4  Second  2   True
#5  Second  2   True

If you need to turn the first values of every group to None, you can do this:

df.loc[df.groupby('A').head(1).index,'C'] = None
        A  B    C
#0   First  1  NaN
#1   First  2  0.0
#2   First  3  0.0
#3  Second  2  NaN
#4  Second  2  1.0
#5  Second  2  1.0

Upvotes: 2

Related Questions