如何在一个to_excel（）和一个read_excel（）之间保留一个大 pandas 多索引？ [英] How can I preserve a pandas multi-index between a to_excel() and a read_excel()?

问题描述

根据 read_excel

的大熊猫文档， / a>，我可以将索引列名称放在一行上，然后该方法将哪些列用作索引。

我想从多索引的数据帧创建一个可以读取的Excel文件，但是我无法弄清楚如何获取大熊猫写的 to_excel ，以便创建此附加行（来自多索引数据帧）。

I want to create an Excel file from a multi-indexed dataframe that can be read in as such, but I can't figure out how to get pandas to write to_excel in such a way that this additional row is created (from a multi-indexed dataframe).

我无法想象将多索引数据框存储为Excel工作表，然后在以后将其拉回来是不常见的一个用例，所以我想知道如果我只是避开'想知道如何做到这一点。

I can't imagine that storing a multi-indexed dataframe as an Excel worksheet and then pulling it back in later is that uncommon a use case, so I'm wondering if I just haven't figured out how to do this.

这是一个数据框的例子，我想在readi之前在冻结Excel不需要告诉read_excel哪些列是索引：

Here's an example of a dataframe I'd like to 'freeze' in Excel before reading back in without having to tell read_excel which columns are the indices:

ipdb> my_df
                             Date         Amount
Rec Section             Row                         
0   Top Section         2    2015-05-01      -105.00
1   Middle Section      3    2015-05-04     90247.60
2   Middle Section      4    2015-05-05     -2992.99
3   Bottom Section      5    2015-05-08      -800.00

在我的例子中，有三个索引列：Rec，Section和Row。

In my example, there are three index columns: Rec, Section, and Row.

当我将其写入Excel然后阅读它时，不想说这个。因为read_excel似乎有一个方法，当它们出现在一个单独的行上时会推断索引名称，所以我想让它只是想出来（假设我正确地写了Excel文件）。

When I write this to Excel and then read it back in, I don't want to have to tell it this. Since read_excel seems to have a method that infers the index names when they appear on a separate row, I want to have it just figure it out (assuming I correctly write the Excel file).

我缺少什么？

推荐答案

当尝试将数据透视表写入Excel时，遇到同样的问题。我可以通过修改 ../ pandas / core 中的 frame.py 文件来使其工作。如果self.columns.nlevels>更改 1 to if self.columns.nlevels>

I was encountering the same issue when trying to write a pivot table to Excel. I was able to get this to work by modifying the frame.py file in ../pandas/core. Changing if self.columns.nlevels > 1 to if self.columns.nlevels > 1 and not index got me what I needed.

由于此功能仍然不支持熊猫，你可能还会遇到有趣的输出。此外，这也可能不会解决 read_excel 的问题。希望这有一点帮助！

As this functionality is still not supported by Pandas, you may still encounter funny output. Also, this will likely not solve the issue for read_excel either. Hopefully this helps a little!

我在他的 GitHub评论。

这篇关于如何在一个to_excel（）和一个read_excel（）之间保留一个大 pandas 多索引？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！