重命名数据透视表中的列 [英] Renaming Columns from pivot_table

问题描述

我在数据帧上运行pivot_table，结果列名称为:

I run a pivot_table on a dataframe and the resulting column names are:

('firstvar', 8001)    ('firstvar', 8002)    ('secondvar', 8001)    ('secondvar', 8002)

但是在使用数据之前，我需要将这些列重命名为以下几行:

But before I can use the data I need to rename the columns to something along these lines:

firstvar_8001    firstvar_8002    secondvar_8001    secondvar_8002

我尝试查看是否可以在执行数据透视表时完成此操作，但我认为这不可能. 因此，相反，我正在寻找一种解决方案，以便在将数据帧中的列保存为csv之前重命名

I tried to see if this could be done when doing the pivot table but I don't think it can be. So instead I'm looking for a solution to renaming the columns in the dataframe before I save it as a csv

我需要在可能更改8000代码的多个文件上运行此文件，因此我考虑使用regex，但找不到使用regex重命名数据框中的列的任何支持.

I need to run this on multiple files where the 8000 codes might change so I was thinking of using regex but I can't find any support for renaming columns in a dataframe using regex.

所以我现在意识到名称实际上是python tuples，我正在努力将其转换为字符串并立即加入它们.

So I realize now that the names are actually python tuples and I'm working on converting them to strings and joining them now.

所以我发现了一个笨拙的解决方案.我的问题是不允许我重命名元组，所以我最终只是导出并重新导入以平整级别，然后以这种方式重命名它们.我真的不喜欢这种解决方案，因此，如果有人提出了更好的解决方案的建议，那将是一个好主意.我真的很感激！

So I found a clumsy solution. My issue was that it wouldn't allow me to rename the tuples so I ended up just exporting and re-import to flatten the levels and then renaming them in that fashion. I really don't like the solution though so if anyone has a suggestion for a better way to do it; I would really appreciate it!

推荐答案

最新答案，但可能对其他人有帮助...

Late answer but can be helpful for others...

my_colnames = []
for i, a in enumerate(my_pivoted_table.columns.get_level_values(1)):
    my_colnames.append(str(a) + '_' + str(my_pivoted_table.columns.get_level_values(0)[i]))
my_pivoted_table.columns = my_colnames

这篇关于重命名数据透视表中的列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！