You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have delved deeper into hyperparameter tuning in pycaret (time series). Currently, it seems that only random_search and grid_search are available as search strategies for time series.
random_search is suboptimal since it is a random (unsystematic) search.
grid_search is also suboptimal because it involves complete (unsystematic) enumeration.
Is there currently any option or workaround for the time series case in Pycaret to perform a proper hyperparameter tuning (e.g., using genetic algorithms which are part of Optuna)?
If it's currently not possible, is there a plan for such a feature (e.g., integration of Optuna into Pycaret for time series forecasting) in the future? If yes, when is this planned?
Pycaret really is a great package. In my opinion, however, it is really sad that this important functionality seems not yet available.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Hello,
I have delved deeper into hyperparameter tuning in pycaret (time series). Currently, it seems that only random_search and grid_search are available as search strategies for time series.
random_search is suboptimal since it is a random (unsystematic) search.
grid_search is also suboptimal because it involves complete (unsystematic) enumeration.
Is there currently any option or workaround for the time series case in Pycaret to perform a proper hyperparameter tuning (e.g., using genetic algorithms which are part of Optuna)?
If it's currently not possible, is there a plan for such a feature (e.g., integration of Optuna into Pycaret for time series forecasting) in the future? If yes, when is this planned?
Pycaret really is a great package. In my opinion, however, it is really sad that this important functionality seems not yet available.
Beta Was this translation helpful? Give feedback.
All reactions