Pandas how to apply a function that takes two arguments-CodePudding

I have a pandas dataframe that contains a date column where the dates are stored as strings:

0    2021-12-04
1    2021-12-01
2    2021-11-29
3    2021-11-15
4    2021-11-06
Name: date, dtype: object

I have a solution that uses variable assignment:

df['date'] = df.apply(lambda x: datetime.strptime(x['date'], '%Y-%m-%d'), axis=1)

But since this transformation is part of a data pipeline, I want to use the assign method. I tried:

df.assign(date=df['date'].apply(datetime.strptime('%Y-%m-%d')))

But this produces an error: KeyError: 'date'.

I suspect this is because the values from the date column aren't being passed to datetime.strptime('%Y-%m-%d'). What is the best way to solve this error?

CodePudding user response：

Use pd.to_datetime instead of apply lambda datetime.strptime:

df.assign(date=pd.to_datetime(df['date'], format='%Y-%m-%d'))