重新采样Pandas数据框并合并列中的字符串 [英] resample Pandas dataframe and merge strings in column

问题描述

我想对熊猫数据框重新采样，并将不同的功能应用于不同的列.问题是我无法正确处理带有字符串的列.我想应用一个将字符串与定界符(例如-")合并的函数.这是一个数据示例:

I want to resample a pandas dataframe and apply different functions to different columns. The problem is that I cannot properly process a column with strings. I would like to apply a function that merges the string with a delimiter such as " - ". This is a data example:

import pandas as pd
import numpy as np
idx = pd.date_range('2017-01-31', '2017-02-03')
data=list([[1,10,"ok"],[2,20,"merge"],[3,30,"us"]])
dates=pd.DatetimeIndex(['2017-01-31','2017-02-03','2017-02-03'])
d=pd.DataFrame(data, index=,columns=list('ABC'))

            A   B          C
2017-01-31  1  10         ok
2017-02-03  2  20      merge
2017-02-03  3  30         us 

使用求和和平均值聚合器对数字列A和B重新采样.但是C列对sum起作用(但它排在第二位，这可能意味着某些地方失败了).

Resampling the numeric columns A and B with a sum and mean aggregator works. Column C however kind of works with sum (but it gets placed on the second place, which might mean that something fails).

d.resample('D').agg({'A': sum, 'B': np.mean, 'C': sum})

              A               C     B
2017-01-31  1.0               a  10.0
2017-02-01  NaN               0   NaN
2017-02-02  NaN               0   NaN
2017-02-03  5.0        merge us  25.0

我想得到这个:

...
2017-02-03  5.0      merge - us  25.0

我尝试以不同的方式使用lambda，但没有成功(未显示).

I tried using lambda in different ways but without success (not shown).

如果我要问另一个相关的问题:我可以对此做一些后期处理，但是如何用零或"填充不同列中的缺失单元格呢?

If I may ask a second related question: I can do some post processing for this, but how to fill missing cells in different columns with zeros or ""?

推荐答案

您的列'C'的agg函数应为join

Your agg function for column 'C' should be a join

d.resample('D').agg({'A': sum, 'B': np.mean, 'C': ' - '.join})

              A     B           C
2017-01-31  1.0  10.0          ok
2017-02-01  NaN   NaN            
2017-02-02  NaN   NaN            
2017-02-03  5.0  25.0  merge - us

这篇关于重新采样Pandas数据框并合并列中的字符串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！