-
|
Like in here: autogluon/autogluon#3529 Which uses this: https://www.cs.cornell.edu/~caruana/ctp/ct.papers/caruana.icml04.icdm06long.pdf |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 3 replies
-
|
@jobs-git, I believe you are looking for https://www.sktime.net/en/latest/api_reference/auto_generated/sktime.forecasting.compose.AutoEnsembleForecaster.html Also see the full API reference with estimators ordered taxonomically: You can also build your own auto-ML algorithms from a few components, see the sectoin "autoML" in the forecasting tutorial. |
Beta Was this translation helpful? Give feedback.
-
|
That is correct, but what algorithm does it use for determining the best weights? Was it based on a certain paper/publication? |
Beta Was this translation helpful? Give feedback.
-
|
Hm, it seems like documentation on that is not present. Would you like to open an issue to add that documentation?
It seems like there are two algorithms:
|
Beta Was this translation helpful? Give feedback.
-
|
Tnx! I created two issues: For documentation update: #8184 New method based on the paper I linked: #8185 |
Beta Was this translation helpful? Give feedback.
@jobs-git, I believe you are looking for https://www.sktime.net/en/latest/api_reference/auto_generated/sktime.forecasting.compose.AutoEnsembleForecaster.html
Also see the full API reference with estimators ordered taxonomically:
https://www.sktime.net/en/stable/api_reference/forecasting.html
You can also build your own auto-ML algorithms from a few components, see the sectoin "autoML" in the forecasting tutorial.