python 分组累加

import pandas as pd
import numpy as np

df = pd.DataFrame.from_dict({'id': ['A', 'B', 'A', 'C', 'C', 'B', 'C'], 'val': [1,2,3,1,5,6,2], 'stuff':['12','23232','13','1234','3235','3236','732323']})
print(df)
df['cumsum']=df.groupby(['id'])['val'].cumsum()
print(df.sort_values(['id']))`


输出结果

  id  val   stuff
0  A    1      12
1  B    2   23232
2  A    3      13
3  C    1    1234
4  C    5    3235
5  B    6    3236
6  C    2  732323

  id  val   stuff  cumsum
0  A    1      12       1
2  A    3      13       4
1  B    2   23232       2
5  B    6    3236       8
3  C    1    1234       1
4  C    5    3235       6
6  C    2  732323       8