```unique_id ds cutoff y Linear
0 AirPassengers 19...
# mlforecastm
Ml Club
07/19/2024, 4:41 PM Copy code
unique_id ds cutoff y Linear
0 AirPassengers 1956-01-01 1955-12-01 284 286.276733
1 AirPassengers 1956-02-01 1955-12-01 277 286.276733
2 AirPassengers 1956-03-01 1955-12-01 317 286.276733
3 AirPassengers 1956-04-01 1955-12-01 313 286.276733
4 AirPassengers 1956-05-01 1955-12-01 318 286.276733j
José Morales
07/19/2024, 4:46 PMYou have to provide
static_features=[]m
Ml Club
07/19/2024, 5:36 PMi have removed the
static_features=[]j
José Morales
07/19/2024, 5:37 PMfit and cross validation are different things. And they both need you to set that argument, not removing it
m
Ml Club
07/19/2024, 5:44 PMi want to Evaluate my model using the
backward trendMl Club
07/19/2024, 5:46 PMLike e.g. :
MSE(df[trend]-df[backward trend])j
José Morales
07/19/2024, 5:53 PMwhat is the backward trend?
m
Ml Club
07/19/2024, 6:31 PMi mean the fitted value,
df_backward = fcst.forecast_fitted_values()j
José Morales
07/19/2024, 6:32 PMso the training error?
m
Ml Club
07/19/2024, 6:32 PMi want to evaluate my model using the forecast_fitted_values data. is that possible?
j
José Morales
07/19/2024, 6:33 PMwell you have all of the inputs, so you should be able to do it in the same way you've done it in the past