Consider the time column in the time slice method

The signature of the time_slice contains a time_column , which is ignored and it's just assumed that the dataframe contains a DateTimeIndex.
Check should be so that if there is no DateTimeIndex, there should be one created, using that column.

def time_slice(df: pd.DataFrame, time_column:str, start: datetime.datetime = None, end: datetime.datetime = None):

I think this is already taken into account here:

arcus.ml/arcus/ml/timeseries/timeops.py

Line 51 in 259b071

if(df.index is None or type(df.index) != pd.DatetimeIndex):

.

Agreed.