代码之家 › 专栏 › 技术社区 › Chris

python-将多个字符串datetimes格式转换为特定的日期格式

datetime date pandas python

Chris · 技术社区 · 4 年前

df = pd.DataFrame({'date': {0: '2020-03-21T10:13:08',  1: '2020-03-21T17:43:03',  2: '2020-03-21T13:13:30',  3: '2020-03-21T20:43:02',  4: '3/8/20 5:31',  5: '3/8/20 5:19',  6: '3/22/20 23:45',  7: '3/22/20 23:45',  8: '2/1/2020 11:53',  9: '2/1/2020 10:53',  10: '1/31/2020 15:20',  11: '1/31/2020 10:37',  12: '2020-04-04 23:34:21',  13: '2020-04-04 23:34:21'}}, 
             index=range(0,14))

我需要将所有这些不同的日期时间字符串转换为日期格式。我使用的方法是:

找到第一个空格并提取日期
在给定特定字符串长度的情况下更改其格式(每个字符串长度都有其特定的日期格式,如下所示 format 参数)
df .

您可以在这里看到这种方法:

df.loc[df["date"].str.find(" ") == 10, "date"] = pd.to_datetime(df.loc[df["date"].str.find(" ") == 10, "date"].str[0:10])
df.loc[df["date"].str.find(" ") == -1, "date"] = pd.to_datetime(df.loc[df["date"].str.find(" ") == 10, "date"].str[0:10])
df.loc[df["date"].str.find(" ") == 6, "date"] = pd.to_datetime(df.loc[df["date"].str.find(" ") == 6, "date"].str[0:6], format="%m/%d/%y")
df.loc[df["date"].str.find(" ") == 7, "date"] = pd.to_datetime(df.loc[df["date"].str.find(" ") == 7, "date"].str[0:7], format="%m/%d/%y")
df.loc[df["date"].str.find(" ") == 8, "date"] = pd.to_datetime(df.loc[df["date"].str.find(" ") == 8, "date"].str[0:8], format="%m/%d/%Y")
df.loc[df["date"].str.find(" ") == 9, "date"] = pd.to_datetime(df.loc[df["date"].str.find(" ") == 9, "date"].str[0:9], format="%m/%d/%Y")

顺便说一下,它必须是可伸缩的(每个格式字符串有很多行)

2 回复 | 直到 4 年前

jezrael 4 年前

对于我来说,将所有值转换为datetimes,然后使用 Series.dt.floor Series.dt.date 如果输出是python日期:

df['date'] = pd.to_datetime(df['date']).dt.floor('d')
#dates
#df['date'] = pd.to_datetime(df['date']).dt.date
print (df)
         date
0  2020-03-21
1  2020-03-21
2  2020-03-21
3  2020-03-21
4  2020-03-08
5  2020-03-08
6  2020-03-22
7  2020-03-22
8  2020-02-01
9  2020-02-01
10 2020-01-31
11 2020-01-31
12 2020-04-04
13 2020-04-04

df['date'] = pd.to_datetime(df['date'].str[:10].str.split().str[0])

Gustavo Gradvohl 4 年前

import pandas as pd

df = pd.DataFrame({'date': {0: '2020-03-21T10:13:08',  1: '2020-03-21T17:43:03',  2: '2020-03-21T13:13:30',  3: '2020-03-21T20:43:02',  4: '3/8/20 5:31',  5: '3/8/20 5:19',  6: '3/22/20 23:45',  7: '3/22/20 23:45',  8: '2/1/2020 11:53',  9: '2/1/2020 10:53',  10: '1/31/2020 15:20',  11: '1/31/2020 10:37',  12: '2020-04-04 23:34:21',  13: '2020-04-04 23:34:21'}}, 
             index=range(0,14))
df
    date
0   2020-03-21T10:13:08
1   2020-03-21T17:43:03
2   2020-03-21T13:13:30
3   2020-03-21T20:43:02
4   3/8/20 5:31
5   3/8/20 5:19
6   3/22/20 23:45
7   3/22/20 23:45
8   2/1/2020 11:53
9   2/1/2020 10:53
10  1/31/2020 15:20
11  1/31/2020 10:37
12  2020-04-04 23:34:21
13  2020-04-04 23:34:21

df['date'] = pd.to_datetime(df['date'])
df
    date
0   2020-03-21 10:13:08
1   2020-03-21 17:43:03
2   2020-03-21 13:13:30
3   2020-03-21 20:43:02
4   2020-03-08 05:31:00
5   2020-03-08 05:19:00
6   2020-03-22 23:45:00
7   2020-03-22 23:45:00
8   2020-02-01 11:53:00
9   2020-02-01 10:53:00
10  2020-01-31 15:20:00
11  2020-01-31 10:37:00
12  2020-04-04 23:34:21
13  2020-04-04 23:34:21