time_varying_media and calibration

[malitsadok1]

I am working on adding a lift test to my MMM model and enabling time_varying_media=True.

For the lift test, I used the method mmm.add_lift_test_measurements(df_lift_test) and added the date column to df_lift_test in two different formats:

1. "date": [pd.to_datetime("2024-08-01")]
2. "date": ["2024-08-01"]

Despite trying both formats, I am still encountering the following error:

MissingValueError                         Traceback (most recent call last)
<cell line: 1>()
----> 1 mmm.add_lift_test_measurements(df_lift_test)

3 frames
/usr/local/lib/python3.10/dist-packages/pymc_marketing/mmm/lift_test.py in assert_is_subset(required, available)
    248     missing = required - available
    249     if missing:
--> 250         raise MissingValueError(list(missing), list(required))
    251 
    252 

MissingValueError: "The value ['date'] is missing from the required ['channel', 'date']"

Thank you in advance for your assistance!

[wd60622]

Ive indentified that the df_lift_test_scaled no longer has the date column which causes error. This needs to either be kept while scaling or added back on in case of time varying media

Would you want to make a PR @malitsadok1 ?

[malitsadok1]

sure