CODEWITHSUNDEEP
python-3.xpandasdatetime
duckman
duckman

Reputation: 747

compare pandas column to a datetime variable

I have a pandas data frame with a column containing datetime values. I used pd.to_datetime() to convert the values to datetime objects. I want to compare these values to a single datetime variable such as TODAY. I got TODAY from datetime.date.today(). Here is my code

data['date'] = pd.to_datetime(data['date'])
data['choose'] =  data['date'] > datetime.date.today()

However, this does not work and I got this error:

TypeError: Invalid comparison between dtype=datetime64[ns] and date

I got another way which is to create a column that contains TODAY for every single row. 

data['today'] = datetime.date.today() 
data['today'] = pd.to_datetime(data['today'])
data['choose'] =  data['date'] > data['today']

But this is inefficient as it takes up memory as it creates another column. What would be the most efficient way to achieve this?

Upvotes: 0

Views: 1115

Answers (1)

Trenton McKinney
Trenton McKinney

Reputation: 62383

  • .dt.date will convert the dataframe series to datetime.date, which can be compared to date.today().
import pandas as pd
from datetime import date, datetime

# setup dataframe
data = {'date': pd.bdate_range(datetime.today(), periods=15).tolist()}
df = pd.DataFrame(data)

# Boolean
df['choose'] = df['date'].dt.date > date.today()

print(df)

      date  choose
2020-05-04   False
2020-05-05    True
2020-05-06    True
2020-05-07    True
2020-05-08    True
2020-05-11    True
2020-05-12    True
2020-05-13    True
2020-05-14    True
2020-05-15    True
2020-05-18    True
2020-05-19    True
2020-05-20    True
2020-05-21    True
2020-05-22    True

Upvotes: 1

Related Questions