熊猫:从非常规时间戳中提取日期 - YYY/MM/DD:HH:MM:SS AM

Question

>我有以下格式的数据：

d = {'Name': ['bob', 'george'], 'date1': ['2014/05/12:12:00:00 AM', '2014/05/15:12:00:00 AM'], 'date2': ['2014/08/31:11:59:59 PM', '2014/08/12:11:59:59 PM']}
df = pd.DataFrame(data=d)
Name                   date1                   date2
0     bob  2014/05/12:12:00:00 AM  2014/08/31:11:59:59 PM
1  george  2014/05/15:12:00:00 AM  2014/08/12:11:59:59 PM

如何操作两个日期列以获得以下格式？

预期产出：

Name       date1       date2
0     bob  2014-05-12  2014-08-31
1  george  2014-05-15  2014-08-12

Answer 1

将pd.to_datetime与format一起使用：

In [1177]: df.date1 = pd.to_datetime(df.date1, format='%Y/%m/%d:%I:%M:%S %p').dt.date     
In [1178]: df.date2 = pd.to_datetime(df.date2, format='%Y/%m/%d:%I:%M:%S %p').dt.date  
In [1179]: df    
Out[1179]: 
Name       date1       date2
0     bob  2014-05-12  2014-08-31
1  george  2014-05-15  2014-08-12

Answer 2

将pandas.to_datetime与format参数一起使用：

df['date1'] = pandas.to_datetime(df['date1'], format='%Y/%m/%d:%H:%M:%S %p')

熊猫文档：https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.to_datetime.html

格式化文档：https://docs.python.org/3/library/datetime.html#strftime-and-strptime-behavior

Answer 3

你可以做：

dates = ['date1','date2']
df[dates] = df[dates].apply(lambda x:pd.to_datetime(x).dt.normalize())

输出：

Name      date1      date2
0     bob 2014-05-12 2014-08-31
1  george 2014-05-15 2014-08-12

Answer 4

试试这个：

d = {'Name': ['bob', 'george'], 'date1': ['2014/05/12:12:00:00 AM', '2014/05/15:12:00:00 AM'], 'date2': ['2014/08/31:11:59:59 PM', '2014/08/12:11:59:59 PM']}
df = pd.DataFrame(data=d)
df['date1'] = pd.to_datetime(df['date1'], format='%Y/%m/%d:%H:%M:%S %p').dt.date
df['date2'] = pd.to_datetime(df['date2'], format='%Y/%m/%d:%H:%M:%S %p').dt.date
print(df)
Name       date1       date2
0     bob  2014-05-12  2014-08-31
1  george  2014-05-15  2014-08-12

Answer 5

你来了：

import pandas as pd
from datetime import datetime
d = {'Name': ['bob', 'george'], 'date1': ['2014/05/12:12:00:00 AM', '2014/05/15:12:00:00 AM'], 'date2': ['2014/08/31:11:59:59 PM', '2014/08/12:11:59:59 AM']}
df = pd.DataFrame(data=d)
date_extractor = lambda dts : datetime.strptime(dts, '%Y/%m/%d:%I:%M:%S %p').date()
df['date1'] = df['date1'].apply(date_extractor)
df['date2'] = df['date2'].apply(date_extractor)
print(df)

输出：

Name       date1       date2
0     bob  2014-05-12  2014-08-31
1  george  2014-05-15  2014-08-12