familiarHyperparameterLearner-class {familiar}

Hyperparameter learner.

Description

A familiarHyperparameterLearner object is a self-contained model that can be applied to predict optimisation scores for a set of hyperparameters.

Details

Hyperparameter learners are used to infer the optimisation score for sets of hyperparameters. These are then used to either infer utility using acquisition functions or to generate summary scores to identify the optimal model.

Slots

name

Name of the familiarHyperparameterLearner object.

learner

Algorithm used to create the hyperparameter learner.

target_learner

Algorithm for which the hyperparameters are being learned.

target_outcome_type

Outcome type of the learner for which hyperparameters are being modeled. Used to determine the target hyperparameters.

optimisation_metric

One or metrics used to generate the optimisation score.

optimisation_function

Function used to generate the optimisation score.

model

The actual model trained using the specific algorithm, e.g. a isolation forest from the isotree package.

target_hyperparameters

The names of the hyperparameters that are used to train the hyperparameter learner.

project_id

Identifier of the project that generated the familiarHyperparameterLearner object.

familiar_version

Version of the familiar package.

package

Name of package(s) required to executed the hyperparameter learner itself, e.g. laGP.

package_version

Version of the packages mentioned in the package attribute.