search

robjhyndman / forecast

Forecasting Functions for Time Series and Linear Models
http://pkg.robjhyndman.com/forecast
1.13k stars 342 forks source link

Suggestion: auto.arima option "save_models = FALSE" #840

Closed sbcalaff closed 4 years ago

sbcalaff commented 4 years ago

In order to save RAM memory, I was wondering if it would be possible to add an option to forecast::auto.arima that makes not saving the models that have been tested. For example, something like save_models = FALSE. This option could be TRUE by default, but if many models are tested, RAM could be saved with a FALSE value.

robjhyndman commented 4 years ago

You can easily remove any elements you do not want to save. e.g., if fc is your forecast object, fc$model ← NULL will remove the model component.

Rob J Hyndman robjhyndman.com ( https://robjhyndman.com )

On Sat, Feb 08, 2020 at 21:00:35, sbcalaff < notifications@github.com > wrote:

In order to save RAM memory, I was wondering if it would be possible to add an option to forecast::auto.arima that makes not saving the models that have been tested. For example, something like save_models = FALSE. This option could be TRUE by default, but if many models are tested, RAM could be saved with a FALSE value.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub ( https://github.com/robjhyndman/forecast/issues/840?email_source=notifications&email_token=AAA7EHUAUWRCV7BOUIDGCDLRBZ7EHA5CNFSM4KRZQJHKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IL7NIPA ) , or unsubscribe ( https://github.com/notifications/unsubscribe-auth/AAA7EHW5FAXJI6NJMRIPN4TRBZ7EHANCNFSM4KRZQJHA ).