除一只外，所有的 pandas 行总和 [英] All row sum with pandas except one

问题描述

我在PostgreSQL数据库上有几个表，这些表看起来或多或少像这样:

I have several tables on a PostgreSQL database that look more or less like that:

gid      col2       col1        col3
6        15         45          77
1        15         45          57
2        14         0.2         42
3        12         6           37
4        9          85          27
5        5          1           15

对于每个表，数字和列的名称都会更改(我在python的循环中创建了它们).

For each table, numbers and columns' names change (I created them in a loop in python).

我想为每个表返回另一列，称为sum，其中包含除gid之外的每个calumn的总和.目标是拥有类似的东西:

I would like to have back another column called sum for each table with the sum of each calumn except for the gid. The goal is having something like that:

gid     col2       col1        col3     sum 
6        15         45          77      137
1        15         45          57      117
2        14         0.2         42      56.2
3        12         6           37      55
4        9          85          27      121 
5        5          1           15      21

我不能使用列名:唯一未更改的列是gid.

I cannot use column name: the only one with no changes is gid.

有人想用python(pandas，numpy)或psql做成它吗?

Some idea to make it with python (pandas, numpy) or psql?

推荐答案

使用 drop + sum :

df['sum'] = df.drop('gid', axis=1).sum(axis=1)
print (df)
   gid  col2  col1  col3    sum
0    6    15  45.0    77  137.0
1    1    15  45.0    57  117.0
2    2    14   0.2    42   56.2
3    3    12   6.0    37   55.0
4    4     9  85.0    27  121.0
5    5     5   1.0    15   21.0

如果gid始终是第一列，请通过 iloc 所有列，而无需先sum:

If gid is always first column, select by iloc all columns without first and then sum them:

df['sum'] = df.iloc[:, 1:].sum(axis=1)
print (df)
   gid  col2  col1  col3    sum
0    6    15  45.0    77  137.0
1    1    15  45.0    57  117.0
2    2    14   0.2    42   56.2
3    3    12   6.0    37   55.0
4    4     9  85.0    27  121.0
5    5     5   1.0    15   21.0

这篇关于除一只外，所有的 pandas 行总和的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！