根据日期和事件计算持续时间

Question

我有一个DF，我想计算收益日期之间的时间。

DF

Date          Earnings Reported           
2018-04-02    1          
2018-04-03    0
2018-04-04    0

DF - 所需

Date          Earnings Reported       DaySinceEarnings      
2018-04-02    1                       0
2018-04-03    0                       1
2018-04-04    0                       2 

我尝试做一个lambda函数，但无法正常工作

df['DaySinceEarnings'] = df.groupby['Earnings Reported'].apply(lambda x: (x == '1') * (x == '1').cumsum())

Answer 1

import pandas as pd

df = pd.DataFrame(
    {'Date': ['2018-04-02',
              '2018-04-03',
              '2018-04-04',
              '2018-04-05',
              '2018-04-06',
              '2018-04-07', ],
     'Earnings Reported': [1, 0, 0, 1, 1, 0]}
)

df['Date'] = pd.to_datetime(df['Date'])


def only_include_reported_days(x):
    x['DaySinceEarnings'] = 0
    if x['Earnings Reported'] == 1:
        return x

    sub = df[(df['Date'] < x['Date']) &
             (df['Earnings Reported'] == 1)]

    x['DaySinceEarnings'] = (x['Date'] - max(sub['Date'])).days
    return x