My objective is simple but not sure if it's possible. Reproducible example:
Can you go from this:
raw_data = {'score': [1, 3, 4, 4, 1, 2, 2, 4, 4, 2],
'player': ['Miller', 'Jacobson', 'Ali', 'George', 'Cooze', 'Wilkinson', 'Lewis', 'Lewis', 'Lewis', 'Jacobson']}
df = pd.DataFrame(raw_data, columns = ['score', 'player'])
df
score player
0 1 Miller
1 3 Jacobson
2 4 Ali
3 4 George
4 1 Cooze
5 2 Wilkinson
6 2 Lewis
7 4 Lewis
8 4 Lewis
9 2 Jacobson
To this:
score col_1 col_2 col_3 col_4
score
1 2 Miller Cooze n/a n/a
2 3 Wilkinson Lewis Jacobson n/a
3 1 Jacobson n/a n/a n/a
4 4 Ali George Lewis Lewis
Via a groupby
?
I can get this far df.groupby(['score']).agg({'score': np.size})
but can't work out how to create the new columns with the column values.
I can duplicate your output with
Option 1
g = df.groupby('score').player
g.size().to_frame('score').join(g.apply(list).apply(pd.Series).add_prefix('col_'))
score col_0 col_1 col_2 col_3
score
1 2 Miller Cooze NaN NaN
2 3 Wilkinson Lewis Jacobson NaN
3 1 Jacobson NaN NaN NaN
4 4 Ali George Lewis Lewis
Option 2
d1 = df.groupby('score').agg({'score': 'size', 'player': lambda x: tuple(x)})
d1.join(pd.DataFrame(d1.pop('player').values.tolist()).add_prefix('col_'))
score col_0 col_1 col_2 col_3
score
1 2 Miller Cooze NaN NaN
2 3 Wilkinson Lewis Jacobson NaN
3 1 Jacobson NaN NaN NaN
4 4 Ali George Lewis Lewis
The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.