MultiplexRegressor
MultiplexRegressor for selecting among different models.
MultiplexRegressor facilitates a framework for performing model selection process over different model classes. It should be used in conjunction with GridSearchCV to get full utilization. It can be used with univariate and multivariate regressors, single-output and multi-output regressors.
MultiplexRegressor is specified with a (named) list of regressors and a selected_regressor hyper-parameter, which is one of the regressor names. The MultiplexRegressor then behaves precisely as the regressor with name selected_regressor, ignoring functionality in the other regressors.
When used with GridSearchCV, MultiplexRegressor provides an ability to tune across multiple estimators, i.e., to perform AutoML, by tuning the selected_regressor hyper-parameter. This combination will then select one of the passed regressors via the tuning algorithm.
Quickstart
from sktime.regression.compose import MultiplexRegressor
estimator = MultiplexRegressor(regressors: list, selected_regressor=None)Parameters(2)
- regressorslist of sktime regressors, or
- list of tuples (str, estimator) of sktime regressors MultiplexRegressor can switch (“multiplex”) between these regressors. These are “blueprint” regressors, states do not change when fit is called.
- selected_regressor: str or None, optional, Default=None.
- If str, must be one of the regressor names.
If no names are provided, must coincide with auto-generated name strings. To inspect auto-generated name strings, call get_params.