pandas 如何将行分组为不同的时间段? [英] Panda how to groupby rows into different time buckets?

问题描述

我有一个带有日期时间类型列的数据框，称为时间戳，我想根据时间戳将时间部分分为几个数据框，每个数据框都包含按x分钟数取值的行，其中x是变量.

I have a dataframe with a datetime type column called timestamp, I want to split the dataframe into several dataframes based on timestamp the time part, each dataframe contains rows that value by its value modulo x minutes, where x is a variable.

请注意，e和f的顺序不是原始的.用10分钟为模，我希望所有以3结尾的时间在一起，所有以1结尾的时间在一起，依此类推.

Notice that e and f are not in original order. With modulo 10 minutes, I want all times that end in 3 together, all times that end in 1 together, so on and so forth.

当x = 10时分组

       timestampe            text
0  2016-08-11 12:01:00        a
1  2016-08-13 11:11:00        b
2  2016-08-09 11:13:00        c
3  2016-08-05 11:33:00        d
4  2016-08-19 11:27:00        e
5  2016-08-21 11:43:00        f

进入

       timestampe            text
0  2016-08-11 12:01:00        a
1  2016-08-13 11:11:00        b

0  2016-08-09 11:13:00        c
1  2016-08-05 11:33:00        d
2  2016-08-21 11:43:00        f

0  2016-08-19 11:27:00        e

推荐答案

您的主要工具将是df.timestampe.dt.minute % 10和groupby.
为了方便说明，我使用了apply(pd.DataFrame.reset_index)

Your main tools will be df.timestampe.dt.minute % 10 and groupby.
I used an apply(pd.DataFrame.reset_index) just as a convenience to illustrate

df.groupby(df.timestampe.dt.minute % 10).apply(pd.DataFrame.reset_index)

仅使用groupby也是有利的

for name, group in df.groupby(df.timestampe.dt.minute % 10):
    print
    print(name)
    print(group)

1
           timestampe text
0 2016-08-11 12:01:00    a
1 2016-08-13 11:11:00    b

3
           timestampe text
2 2016-08-09 11:13:00    c
3 2016-08-05 11:33:00    d
5 2016-08-21 11:43:00    f

7
           timestampe text
4 2016-08-19 11:27:00    e

这篇关于 pandas 如何将行分组为不同的时间段?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！