Getting columns with datetime format such as (2017-02-12 10:23:55 AM)[YYYY-MM-dd hh:mm:ss AM/PM] using pandas

Question

I recently asked a question about identifing all the columns which are datetime. Here it is: Get all columns with datetime type using pandas?

The answer was correct for a proper date time format, however, I now realize my data isn't proper date time, it is a string formatted like "2017-02-12 10:23:55 AM" and I was advised to create a new question.

I have a huge dataframe with an unknown number of date time columns, where I do not know their names nor their position. How do I identify the column names of the date time columns which have the date of format such as YYYY-MM-dd hh:mm:ss AM/PM?

So the dates are actually strings? – miradulo Feb 19 '17 at 03:54 — miradulo, Feb 19 '17 at 03:54
Yes, sorry I thought they were proper dates. – J. Doe Feb 19 '17 at 04:01 — J. Doe, Feb 19 '17 at 04:01

score 1 · Accepted Answer · answered Feb 19 '17 at 04:16

1

One way to do this would be to test for successful conversion:

def is_datetime(datetime_string):
    try:
        pd.to_datetime(datetime_string)
        return True
    except ValueError:
        return False

With this:

dt_columns = [c for c in df.columns if is_datetime(df[c][0])]

Note: This tests for any string that can be converted to a datetime.

answered Feb 19 '17 at 04:16

Stephen Rauch

47,830
31
106
135

I see, thank you. I was trying to avoid for loops incase there was a build-in function. But thanks! – J. Doe Feb 19 '17 at 04:19

Getting columns with datetime format such as (2017-02-12 10:23:55 AM)[YYYY-MM-dd hh:mm:ss AM/PM] using pandas

1 Answers1