Function reference • modeltuner

All functions

c(<cv>): Add models to a cv object

crit_min() crit_last() crit_first() crit_iter() crit_se() crit_overfit() crit_list(): Preference criteria for iteratively fitted models

cv() print(<cv>): Run a cross-validation

cv_performance() print(<performance>) plot(<performance>): Calculate train and test errors based on cross-validation.

cv_predict() cv_resid(): Extract out-of-sample predictions and residuals from cross-validation

default_metric(): Get the default metric of an object

evaluation_log() print(<evaluation_log>): Evaluation log

expand_formula(): Expand a formula

extract_fits(): Extract the models fitted in a cross-validation from a “cv” object.

extract_model() extract_multimodel(): Extraction of a model and multimodel

fit() add_fit() has_fit(): Re-fit a model object using the complete model data

fm_const() print(<fm_const>) predict(<fm_const>): Fitting a constant model

fm_glmnet() predict(<fm_glmnet>) coef(<fm_glmnet>) plot(<fm_glmnet>): formula-based wrapper for glmnet()

fm_knn() predict(<fm_knn>): k-Nearest Neighbors model

fm_smooth_spline() predict(<fm_smooth_spline>): Smoothing spline model with formula-interface

fm_xgb() print(<fm_xgb>) predict(<fm_xgb>) extract_booster(): formula-based wrapper for xgb.train()

ifm: Iteratively fitted models (IFM) and preferred iterations

label() `label<-`() set_label() n_model(): Query or set model label(s)

last_cv() set_last_cv(): Get and set the last cv object

make_folds(): Create folds (cross-validation groups)

rmse() mae() medae() mse() logLoss() classification_error(): Metrics

model(<glm>) model(<glmrob>) model(<gam>) model(<ranger>) model(<merMod>) model(<lmerMod>) model(<glmerMod>): Special methods of model()

model() print(<model>): Create a model object

models(): Combine several fitted models in a multimodel

modeltuner-package modeltuner: modeltuner package overview

modeltuner_cheatsheet(): Open a modeltuner cheatsheet

modeltuner_options(): List all options defined in package “modeltuner”

multimodel() print(<multimodel>) c(<model>) c(<multimodel>): Create a multimodel object

null_formula(): “Null formula” of a model

param_table(): Table format with informative print method.

performance(): Evaluate the model performance of a model

plot(<evaluation_log>): Plot method for class “evaluation_log”

plot(<model>) plot(<multimodel>) plot(<cv>): Plot methods for classes “model”, “multimodel” and “cv”

pmodel(): Purely predictive (non-fittable) model

predict(<model>) residuals(<model>) predict(<multimodel>) residuals(<multimodel>): Predictions and residuals from a (multi-)model

response(): Extract the values of the model response from an object

set_metric(): Change the default metric of a cv object

set_pref_iter() extract_pref_iter() expand_pref_iter(): Extract set, and expand preference criteria for a “cv” object for iteratively fitted models

simuldat(): Simulate data

sort_models(): Reorder models in an object of class “multimodel”, “cv”, “performance” or “evaluation_log”

step_extend() step_forward() step_reduce() step_backward() best_subset(): Generate and cross-validate models resulting from adding or removing variables and stepwise procedures

subset(<multimodel>) subset(<cv>) subset(<performance>) subset(<evaluation_log>): Subset an object of class “multimodel”, “cv”, “performance” or “evaluation_log”

tune(): Selection of the best-performing model in a “cv” object

update(<model>) update(<multimodel>) absent() null() unchanged(): Update an object of class “model” or “multimodel”

weights(<model>): Extract the (fitting) weights from a “model” object