如何确定pandas数据中每分钟的num个条目

Question

我试图弄清楚每分钟列出"f"列中的条目的次数.

    import pandas as pd
    import datetime as dt

    f= ['f0001',
        'f0001',
        'f0001',
        'f0001',
        'f0020',
        'f0008',
        'f0001',
        'f0005',
        'f3203',
        'f0002',
        'f0002',
        'f0001',
        'f0201',
        'f0001',
        'f0439',
        'f0233',
        'f0008',
        'f0003',
        'f0009',
        'f0005']


    dates = ['20130101100103', '20130101100110',
               '20130101100125', '20130101100133',
               '20130101100100', '20130101100200',
               '20130101100200', '20130101100200',
               '20130101100200', '20130101100200',
               '20130101100200', '20130101100300',
               '20130101100300', '20130101100300',
               '20130101100300', '20130101100400',
               '20130101100400', '20130101100400',
               '20130101100400', '20130101100400']


    d = {'date':  dates}
    data = pd.DataFrame(d)
    data['user'] = f

    data.date = data.date.apply(str)
    data.date = data.date.apply(lambda x: 
    dt.datetime.strptime(x,'%Y%m%d%H%M%S'))

    s = data.groupby([data.date.map(lambda t: t.minute)]).count()

但到目前为止我所得到的只是下面的内容

s

    date    user
date        
1   5   5
2   6   6
3   4   4
4   5   5

Answer 1

你快到了.您只需要添加data['user']到groupby子句中.

data.groupby([[data.date.dt.minute, data['user']]).count().rename(columns={'date':'count'}).reset_index()

输出:

date    user    count
0   1   f0001   4
1   1   f0020   1
2   2   f0001   1
3   2   f0002   2
4   2   f0005   1
5   2   f0008   1
6   2   f3203   1
7   3   f0001   2
8   3   f0201   1
9   3   f0439   1
10  4   f0003   1
11  4   f0005   1
12  4   f0008   1
13  4   f0009   1
14  4   f0233   1