Right now, I am working with this dataframe..
| Name | DateSolved | Points |
|---|---|---|
| Jimmy | 12/3 | 100 |
| Tim | 12/4 | 50 |
| Jo | 12/5 | 25 |
| Jonny | 12/5 | 25 |
| Jimmy | 12/8 | 10 |
| Tim | 12/8 | 10 |
At this moment, if there are duplicate names in the dataset, I just drop the oldest one (by date) from the dataframe by utilizing df.sort_values('DateSolved').drop_duplicates('Name', keep='last') leading to a dataset like this
| Name | DateSolved | Points |
|---|---|---|
| Jo | 12/5 | 25 |
| Jonny | 12/5 | 25 |
| Jimmy | 12/8 | 10 |
| Tim | 12/8 | 10 |
However, instead of dropping the oldest one, I wish to keep it but give it a 50% points reduction. Something like this
| Name | DateSolved | Points |
|---|---|---|
| Jimmy | 12/3 | 50 (-50%) |
| Tim | 12/4 | 25 (-50%) |
| Jo | 12/5 | 25 |
| Jonny | 12/5 | 25 |
| Jimmy | 12/8 | 10 |
| Tim | 12/8 | 10 |
How could I go about doing this? I cannot find a way to both FIND the duplicates based on "Name" and then change the value of the "POINTS" column in the same row.
Thank you!
CodePudding user response：
IIUC use DataFrame.duplicated for select all duplicates withot last, select column Points and divide by 2:
df.loc[df.duplicated('Name', keep='last'), 'Points'] /= 2
print (df)
Name DateSolved Points
0 Jimmy 12/3 50.0
1 Tim 12/4 25.0
2 Jo 12/5 25.0
3 Jonny 12/5 25.0
4 Jimmy 12/8 10.0
5 Tim 12/8 10.0