Split values in a column in "is a date" or "NaT"

Question

I would like to find values in a column (clear_date) that do not correspond to a valid date. The date is formatted as '%Y/%m/%d'.

I've tried the following piece of code but, the resulting variable doesn't have any rows!

x_test = dataset[dataset['clear_date'] == "NaT"].copy()

Here is a picture for reference:

what's the `dtype` of that column? – Pierre D Jan 24 '21 at 14:38 — Pierre D, Jan 24 '21 at 14:38
@PierreD datetime64[ns] – Ayush Vishwakarma Jan 24 '21 at 14:41 — Ayush Vishwakarma, Jan 24 '21 at 14:41

score 0 · Accepted Answer · answered Jan 24 '21 at 14:42

0

If the clear_date column is already a proper datetime, then:

x_test = dataset.loc[dataset['clear_date'].isnull()].copy()

Otherwise, e.g. if it is a string:

x_test = dataset.loc[pd.to_date_time(dataset['clear_date']).isnull()].copy()

answered Jan 24 '21 at 14:42

Pierre D

The ```clear_date``` column is already a datetime The first code worked for me. Thanks :) – Ayush Vishwakarma Jan 24 '21 at 14:45

1 Answers1