I have a data frame like this:
df:
col1 col2
1 pqr
3 abc
2 pqr
4 xyz
1 pqr
I found that there is duplicate value and its pqr. I want to add 1,2,3 where pqr occurs. The final data frame I want to achieve is:
df1
col1 col2
1 pqr1
3 abc
2 pqr2
4 xyz
1 pqr3
How to do it in efficient way
Use duplicated with keep=False for all dupe rows and add counter created by cumcount:
mask = df['col2'].duplicated(keep=False)
df.loc[mask, 'col2'] += df.groupby('col2').cumcount().add(1).astype(str)
Or:
df['col2'] = np.where(df['col2'].duplicated(keep=False),
df['col2'] + df.groupby('col2').cumcount().add(1).astype(str),
df['col2'])
print (df)
col1 col2
0 1 pqr1
1 3 abc
2 2 pqr2
3 4 xyz
4 1 pqr3
If need same only for pqr values:
mask = df['col2'] == 'pqr'
df.loc[mask, 'col2'] += pd.Series(np.arange(1, mask.sum() + 1),
index=df.index[mask]).astype(str)
print (df)
col1 col2
0 1 pqr1
1 3 abc
2 2 pqr2
3 4 xyz
4 1 pqr3