Pandas:合并两个数据框时控制新列名? [英] Pandas: control new column names when merging two dataframes?

问题描述

我想将两个 Pandas 数据框合并在一起并控制新列值的名称.

I would like to merge two Pandas dataframes together and control the names of the new column values.

我最初从 CSV 文件创建了数据框.原始 CSV 文件如下所示:

I originally created the dataframes from CSV files. The original CSV files looked like this:

   # presents.csv
   org,name,items,spend...
   12A,Clerkenwell,151,435,...
   12B,Liverpool Street,37,212,...
   ...
   # trees.csv
   org,name,items,spend...
   12A,Clerkenwell,0,0,...
   12B,Liverpool Street,2,92,...
   ...

现在我有两个数据框:

df_presents = pd.read_csv(StringIO(presents_txt))
df_trees = pd.read_csv(StringIO(trees_txt))

我想将它们合并在一起以获得最终数据框，加入 org 和 name 值，然后使用适当的前缀为所有其他列添加前缀.

I want to merge them together to get a final data frame, joining on the org and name values, and then prefixing all other columns with an appropriate prefix.

org,name,presents_items,presents_spend,trees_items,trees_spend...
12A,Clerkenwell,151,435,0,0,...
12B,Liverpool Street,37,212,2,92,...

我一直在阅读有关合并和加入的文档.这似乎正确合并并导致正确数量的列:

I've been reading the documentation on merging and joining. This seems to merge correctly and result in the right number of columns:

ad = pd.DataFrame.merge(df_presents, df_trees,
                        on=['practice', 'name'],
                        how='outer')

但是然后执行 print list(aggregate_data.columns.values) 会显示以下列:

But then doing print list(aggregate_data.columns.values) shows me the following columns:

[org', u'name', u'spend_x', u'spend_y', u'items_x', u'items_y'...]

如何将 spend_x 重命名为 presents_spend 等?

How can I rename spend_x to be presents_spend, etc?

推荐答案

合并函数中的 suffixes 选项就是这样做的.默认是后缀=('_x', '_y').

The suffixes option in the merge function does this. The defaults are suffixes=('_x', '_y').

通常，重命名列可以使用 重命名方法.

In general, renaming columns can be done with the rename method.

这篇关于Pandas:合并两个数据框时控制新列名?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！