Advertisement
Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- import pandas as pd
- test_dataframe = pd.DataFrame({'date': ['2020-12-31', '2021-01-02'], 'user_id': [12345, 56789]})
- test_dataframe['date'] = pd.to_datetime(test_dataframe['date'])
- # The following script returns 2021 for the second date but the actual year for that date is needed which is 2020
- test_dataframe['year'] = test_dataframe['date'].dt.year
- test_dataframe['week'] = test_dataframe['date'].dt.week
- test_dataframe['cohort'] = test_dataframe['year'].astype(str) + '-' + test_dataframe['week'].astype(str)
- # The cohort needs to be '2020-53' for both as the data actually belongs to week 53 of the year 2020.
- desired_dataframe = pd.DataFrame({'date': ['2020-12-31', '2021-01-02'], 'user_id': [12345, 56789]
- , 'year': [2020, 2020], 'week': [53, 53], 'cohort': ['2020-53', '2020-53']})
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement