Mixed Timezones DateTime coerce failes

[koschmid]

Hey,

i am currently trying to use a schema to verify typing for some data. In my datetime column i have mixed timezones, represented with offsets. When i try to coerce it failes and it seems like it is because the pd.to_datetime called behind panderas would need to be parametrised with tz=UTC.

My example:

CustomSchema(pa.SchemaModel):
shop_date: Series[DateTime] = pa.Field(coerce=True)

Case 1:
data = DataFrameCustomSchema
-> First prints pandas warning: pd.to_datetime() need tz=UTC for mixed timezones, then panderas throws error.

Case 2:
raw_data["shop_date"] = pd.to_datetime(raw_data["shop_date"], errors="coerce", utc=True)
data = DataFrameCustomSchema
-> Works as supposed to.

I tried solutions from related threads, but nothing worked. Either timezone is completly ignored or i get an error. I want so Datatype Doc, that reads like there is the possibility somewhere under the hood, i probably just don't know how to use it.
https://pandera.readthedocs.io/en/v0.20.2/reference/generated/pandera.engines.pandas_engine.DateTime.html