familiarModel-class: Familiar model.
In familiar: End-to-End Automated Machine Learning and Model Evaluation

familiarModel-class

R Documentation

Familiar model.

Description

A familiarModel object is a self-contained model that can be applied to generate predictions for a dataset. familiarModel objects form the parent class of learner-specific child classes.

Slots

name: Name of the familiarModel object.
model: The actual model trained using a specific algorithm, e.g. a random forest from the ranger package, or a LASSO model from glmnet.
outcome_type: Outcome type of the data used to create the object.
outcome_info: Outcome information object, which contains additional information concerning the outcome, such as class levels.
feature_info: List of objects containing feature information, e.g., name, class levels, transformation, normalisation and clustering parameters.
data_column_info: Data information object containing information regarding identifier column names and outcome column names.
hyperparameters: Set of hyperparameters used to train the model.
hyperparameter_data: Information generated during hyperparameter optimisation.
calibration_model: One or more models used to recalibrate the model output. Currently only used by some models.
novelty_detector: A familiarNoveltyDetector object that can be used to detect out-of-distribution samples.
learner: Learning algorithm used to create the model.
fs_method: Feature selection method used to determine variable importance for the model.
required_features: The set of features required for complete reproduction, i.e. with imputation.
model_features: The set of features that is used to train the model,
novelty_features: The set of features that is used to train all novelty detectors in the ensemble.
calibration_info: Calibration information, e.g. baseline survival in the development cohort.
km_info: Data concerning stratification into risk groups.
run_table: Run table for the data used to train the model. Used internally.
settings: A copy of the evaluation configuration parameters used at model creation. These are used as default parameters when evaluating the model (technically, familiarEnsemble) to create a familiarData object.
is_trimmed: Flag that indicates whether the model, stored in the model slot, has been trimmed.
trimmed_function: List of functions whose output has been captured prior to trimming the model.
messages: List of warning and error messages generated during training.
project_id: Identifier of the project that generated the familiarModel object.
familiar_version: Version of the familiar package.
package: Name of package(s) required to executed the model itself, e.g. ranger or glmnet.
package_version: Version of the packages mentioned in the package attribute.