如果求和列等于零，Groupby .cumsum() 空白？

Question

我有一个 DataFrame .groupby() .cumsum()，其中的 DataFrame 如下：

   Col_A Col_B Col_C
 1   A    0            
 2   A    1     1      
 3   A    1     2      
 4   A    1     3      
 5   B    0     0      
 6   B    1     1      
 7   B    0            
 8   B    1     2      
 9   C    1     1      
10   C    1     2      
11   C    1     3      
12   C    0

Col_B的总和是df.groupby(['Col_A'])['Col_B'].cumsum()。但是，当 Col_B == 0 时，.cumsum() 为空白。如何在 Col_B 为空白时记录 .cumsum()？

生成的 DataFrame 应类似于：

      Col_A Col_B Col_C
     1   A    0     0       
     2   A    1     1      
     3   A    1     2      
     4   A    1     3      
     5   B    0     0      
     6   B    1     1      
     7   B    0     1       
     8   B    1     2      
     9   C    1     1      
    10   C    1     2      
    11   C    1     3      
    12   C    0     3

Answer 1

我想你需要先过滤 boolean indexing or query:

df['Col_C'] = df[df['Col_B'] != 0].groupby(['Col_A'])['Col_B'].cumsum()
print (df)
   Col_A  Col_B  Col_C
1      A      0    NaN
2      A      1    1.0
3      A      1    2.0
4      A      1    3.0
5      B      0    NaN
6      B      1    1.0
7      B      0    NaN
8      B      1    2.0
9      C      1    1.0
10     C      1    2.0
11     C      1    3.0
12     C      0    NaN

或者：

df['Col_C'] = df.query('Col_B != 0').groupby(['Col_A'])['Col_B'].cumsum()
print (df)
   Col_A  Col_B  Col_C
1      A      0    NaN
2      A      1    1.0
3      A      1    2.0
4      A      1    3.0
5      B      0    NaN
6      B      1    1.0
7      B      0    NaN
8      B      1    2.0
9      C      1    1.0
10     C      1    2.0
11     C      1    3.0
12     C      0    NaN

最后用 ffill 替换 NaNs （fillna with method='ffill'). But get first values still NaNs, which are replaced by fillna 最后将列转换为 int:

df['Col_C']  = df['Col_C'].ffill().fillna(0).astype(int)
print (df)
   Col_A  Col_B  Col_C
1      A      0      0
2      A      1      1
3      A      1      2
4      A      1      3
5      B      0      3
6      B      1      1
7      B      0      1
8      B      1      2
9      C      1      1
10     C      1      2
11     C      1      3
12     C      0      3

Answer 2

一列为 0 与一列完全空白不同。如果您在列中有 NA，该列的 .cumsum() 实际上应该是 NA（或如您所说的 'blank' ）。您可以检查整列是否为 NA 并相应地设置值。

Documentation:

DataFrame.cumsum(axis=None, skipna=True, *args, **kwargs)
Return cumulative sum over requested axis.

skipna : boolean, default True
Exclude NA/null values. If an entire row/column is NA, the result will be NA

如果求和列等于零，Groupby .cumsum() 空白？

Groupby .cumsum() blank if the summed column is equal to zero?

python

group-by

sum

series

pandas