How to explode row list into multiple cumulative lists?-CodePudding

Firstly I did a groupby operation: df.groupby('a')['b'].agg(list).reset_index(name='b')

a b
A 1
A 2
B 5
B 5
B 4
C 6

Resulting in this df:

a b
A [1,2]
B [5,5,4]
C [6]

Now I want to explode these lists into multiple cumulative lists by row.

a b
A [1]
A [1,2]
B [5]
B [5,5]
B [5,5,4]
C [6]

CodePudding user response：

You need 1st convert the cell value to list then we can do cumsum

df['out'] = df['b'].apply(lambda x : [x]).groupby(df['a']).apply(lambda x : x.cumsum() )
Out[382]: 
0          [1]
1       [1, 2]
2          [5]
3       [5, 5]
4    [5, 5, 4]
5          [6]
Name: b, dtype: object

CodePudding user response：

As DataFrame.expanding() seems only to work on numeric data, I resort to this nested list comprehension:

df['b'] = [subdf['b'].tolist()[:i 1] 
           for group, subdf in df.groupby('a')
           for i in range(subdf.shape[0])]

print(df)

   a          b
0  A        [1]
1  A     [1, 2]
2  B        [5]
3  B     [5, 5]
4  B  [5, 5, 4]
5  C        [6]