Is there an option to also keep the submodel forecasts in the output when forecasting using an ensemble?
At the moment I'm currently doing something like
# Put both the submodels and ensemble into the table
ensemble_tbl <- submodel_table %>%
add_modeltime_model(ensemble_fit_glmnet)
It works, but is pretty much duplicating the whole forecasting time duration.
I had a look through this repos modeltime_forecast function and I see the forecasts are made but not kept, I was thinking of adding a flag in those forecast functions to keep_submodel_forecasts, what do you think?
Is there an option to also keep the submodel forecasts in the output when forecasting using an ensemble?
At the moment I'm currently doing something like
It works, but is pretty much duplicating the whole forecasting time duration.
I had a look through this repos modeltime_forecast function and I see the forecasts are made but not kept, I was thinking of adding a flag in those forecast functions to
keep_submodel_forecasts
, what do you think?