What's the most efficient way to convert a string like "3:00pm" to a Python datetime-CodePudding

I have a column of times that are stored as strings. What I would like to do is take "3:00pm" and convert it to a datetime object that's 15:00 (24 hour clock).

I could always write a custom function that parses out the string, identifies whether it's am or pm, and create a mapping, but I was wondering if there's any built in functionality in Python or Pandas to do this efficiently.

Thanks so much!

CodePudding user response：

Considering the following dataframe as an example

      Time  Value
0   7:00pm     48
1   8:00pm     48
2  11:00am      8
3   5:00pm      8
4  12:00pm     62
5   8:00am     41
6   9:00am     51
7   7:00pm     70
8   6:00am     41
9   5:00am     72

In order to convert the column Time from 3:00pm to a datetime object that's 15:00 (24 hour clock), one can use pandas.to_datetime as follows

df['Time'] = pd.to_datetime(df['Time'], format = '%I:%M%p')

[Out]:

                 Time  Value
0 1900-01-01 19:00:00     48
1 1900-01-01 20:00:00     48
2 1900-01-01 11:00:00      8
3 1900-01-01 17:00:00      8
4 1900-01-01 12:00:00     62
5 1900-01-01 08:00:00     41
6 1900-01-01 09:00:00     51
7 1900-01-01 19:00:00     70
8 1900-01-01 06:00:00     41
9 1900-01-01 05:00:00     72

Notes:

The format can be adjusted depending on one's goals. See here Python's strftime documentation. For this use case:
- %I: 12-hour format
- %M: minutes
- %p: AM/PM

CodePudding user response：

here is one way to do it

# convert to datetime and gather the time
df['24hr']=pd.to_datetime(df['Time']).dt.time
df

    Time    24hr
0   7:00pm  19:00:00
1   8:00pm  20:00:00
2   11:00am 11:00:00
3   3:00am  03:00:00
4   1:00pm  13:00:00