Function reference • sl3

Defining A Task
`make_sl3_Task()`	Define a Machine Learning Task
`variable_type()`	Specify Variable Type
Finding Learners
`sl3_list_properties()` `sl3_list_learners()`	List sl3 Learners
sl3 Learners
`Lrnr_HarmonicReg`	Harmonic Regression
`Lrnr_arima`	Univariate ARIMA Models
`Lrnr_bartMachine`	bartMachine: Bayesian Additive Regression Trees (BART)
`make_learner()`	Base Class for all sl3 Learners
`Lrnr_bayesglm`	Bayesian Generalized Linear Models
`Lrnr_bilstm`	Bidirectional Long short-term memory Recurrent Neural Network (LSTM)
`Lrnr_bound`	Bound Predictions
`Lrnr_caret`	Caret (Classification and Regression) Training
`Lrnr_cv`	Fit/Predict a learner with Cross Validation
`Lrnr_cv_selector`	Cross-Validated Selector
`Lrnr_dbarts`	Discrete Bayesian Additive Regression Tree sampler
`Lrnr_define_interactions`	Define interactions terms
`Lrnr_density_discretize`	Density from Classification
`Lrnr_density_hse`	Density Estimation With Mean Model and Homoscedastic Errors
`Lrnr_density_semiparametric`	Density Estimation With Mean Model and Homoscedastic Errors
`Lrnr_earth`	Earth: Multivariate Adaptive Regression Splines
`Lrnr_expSmooth`	Exponential Smoothing state space model
`Lrnr_ga`	Nonlinear Optimization via Genetic Algorithm (GA)
`Lrnr_gam`	GAM: Generalized Additive Models
`Lrnr_gbm`	GBM: Generalized Boosted Regression Models
`Lrnr_glm`	Generalized Linear Models
`Lrnr_glm_fast`	Computationally Efficient Generalized Linear Model (GLM) Fitting
`Lrnr_glm_semiparametric`	Semiparametric Generalized Linear Models
`Lrnr_glmnet`	GLMs with Elastic Net Regularization
`Lrnr_glmtree`	Generalized Linear Model Trees
`Lrnr_grf`	Generalized Random Forests Learner
`Lrnr_grfcate`	Generalized Random Forests for Conditional Average Treatment Effects
`Lrnr_gru_keras`	Recurrent Neural Network with Gated Recurrent Unit (GRU) with Keras
`Lrnr_gts`	Grouped Time-Series Forecasting
`define_h2o_X()`	h2o Model Definition
`Lrnr_h2o_grid`	Grid Search Models with h2o
`Lrnr_hal9001`	Scalable Highly Adaptive Lasso (HAL)
`Lrnr_haldensify`	Conditional Density Estimation with the Highly Adaptive LASSO
`Lrnr_hts`	Hierarchical Time-Series Forecasting
`Lrnr_independent_binomial`	Classification from Binomial Regression
`Lrnr_lightgbm`	LightGBM: Light Gradient Boosting Machine
`Lrnr_lstm_keras`	Long short-term memory Recurrent Neural Network (LSTM) with Keras
`Lrnr_mean`	Fitting Intercept Models
`Lrnr_multiple_ts`	Stratify univariable time-series learners by time-series
`Lrnr_multivariate`	Multivariate Learner
`Lrnr_nnet`	Feed-Forward Neural Networks and Multinomial Log-Linear Models
`Lrnr_nnls`	Non-negative Linear Least Squares
`Lrnr_optim`	Optimize Metalearner according to Loss Function using optim
`Lrnr_pca`	Principal Component Analysis and Regression
`Lrnr_polspline`	Polyspline - multivariate adaptive polynomial spline regression (polymars) and polychotomous regression and multiple classification (polyclass)
`Lrnr_pooled_hazards`	Classification from Pooled Hazards
`Lrnr_randomForest`	Random Forests
`Lrnr_ranger`	Ranger: Fast(er) Random Forests
`Lrnr_revere_task`	Learner that chains into a revere task
`Lrnr_rpart`	Learner for Recursive Partitioning and Regression Trees
`Lrnr_rugarch`	Univariate GARCH Models
`Lrnr_screener_augment`	Augmented Covariate Screener
`Lrnr_screener_coefs`	Coefficient Magnitude Screener
`Lrnr_screener_correlation`	Correlation Screening Procedures
`Lrnr_screener_importance`	Variable Importance Screener
`Lrnr_sl`	The Super Learner Algorithm
`Lrnr_solnp`	Nonlinear Optimization via Augmented Lagrange
`Lrnr_solnp_density`	Nonlinear Optimization via Augmented Lagrange
`Lrnr_stratified`	Stratify learner fits by a single variable
`Lrnr_subset_covariates`	Learner with Covariate Subsetting
`Lrnr_svm`	Support Vector Machines
`Lrnr_tsDyn`	Nonlinear Time Series Analysis
`Lrnr_ts_weights`	Time-specific weighting of prediction losses
`Lrnr_xgboost`	xgboost: eXtreme Gradient Boosting
`Lrnr_pkg_SuperLearner`	Use SuperLearner Wrappers, Screeners, and Methods, in sl3
Composing Learners
`Pipeline`	Pipeline (chain) of learners.
`Stack`	Learner Stacking
`customize_chain()`	Customize chaining for a learner
Loss functions
`loss_squared_error()` `loss_loglik_true_cat()` `loss_loglik_binomial()` `loss_loglik_multinomial()` `loss_squared_error_multivariate()`	Loss Function Definitions
`risk()`	Risk Estimation
Metalearner functions
`metalearner_logistic_binomial()` `metalearner_linear()` `metalearner_linear_multivariate()` `metalearner_linear_multinomial()`	Combine predictions from multiple learners
Helpful for Defining Learners
`write_learner_template()`	Generate a file containing a template `sl3` Learner
`args_to_list()`	Get all arguments of parent call (both specified and defaults) as list
`call_with_args()`	Call with filtered argument list
`true_obj_size()`	Estimate object size using serialization
`safe_dim()`	dim that works for vectors too
`delayed_make_learner()` `learner_train()` `delayed_learner_train()` `learner_fit_predict()` `delayed_learner_fit_predict()` `learner_fit_chain()` `delayed_learner_fit_chain()` `learner_subset_covariates()` `learner_process_formula()` `delayed_learner_subset_covariates()` `delayed_learner_process_formula()`	Learner helpers
Sample Datasets
`cpp` `cpp_imputed`	Subset of growth data from the collaborative perinatal project (CPP)
`cpp_1yr`	Subset of growth data from the collaborative perinatal project (CPP)
`bsds`	Bicycle sharing time series dataset
`density_dat`	Simulated data with continuous exposure
Miscellaneous
`sl3Options()`	Querying/setting a single `sl3` option
Index
`customize_chain()`	Customize chaining for a learner
`Lrnr_HarmonicReg`	Harmonic Regression
`Lrnr_arima`	Univariate ARIMA Models
`Lrnr_bartMachine`	bartMachine: Bayesian Additive Regression Trees (BART)
`make_learner()`	Base Class for all sl3 Learners
`Lrnr_bayesglm`	Bayesian Generalized Linear Models
`Lrnr_bilstm`	Bidirectional Long short-term memory Recurrent Neural Network (LSTM)
`Lrnr_bound`	Bound Predictions
`Lrnr_caret`	Caret (Classification and Regression) Training
`Lrnr_cv`	Fit/Predict a learner with Cross Validation
`Lrnr_cv_selector`	Cross-Validated Selector
`Lrnr_dbarts`	Discrete Bayesian Additive Regression Tree sampler
`Lrnr_define_interactions`	Define interactions terms
`Lrnr_density_discretize`	Density from Classification
`Lrnr_density_hse`	Density Estimation With Mean Model and Homoscedastic Errors
`Lrnr_density_semiparametric`	Density Estimation With Mean Model and Homoscedastic Errors
`Lrnr_earth`	Earth: Multivariate Adaptive Regression Splines
`Lrnr_expSmooth`	Exponential Smoothing state space model
`Lrnr_ga`	Nonlinear Optimization via Genetic Algorithm (GA)
`Lrnr_gam`	GAM: Generalized Additive Models
`Lrnr_gbm`	GBM: Generalized Boosted Regression Models
`Lrnr_glm`	Generalized Linear Models
`Lrnr_glm_fast`	Computationally Efficient Generalized Linear Model (GLM) Fitting
`Lrnr_glm_semiparametric`	Semiparametric Generalized Linear Models
`Lrnr_glmnet`	GLMs with Elastic Net Regularization
`Lrnr_glmtree`	Generalized Linear Model Trees
`Lrnr_grf`	Generalized Random Forests Learner
`Lrnr_grfcate`	Generalized Random Forests for Conditional Average Treatment Effects
`Lrnr_gru_keras`	Recurrent Neural Network with Gated Recurrent Unit (GRU) with Keras
`Lrnr_gts`	Grouped Time-Series Forecasting
`define_h2o_X()`	h2o Model Definition
`Lrnr_h2o_grid`	Grid Search Models with h2o
`Lrnr_hal9001`	Scalable Highly Adaptive Lasso (HAL)
`Lrnr_haldensify`	Conditional Density Estimation with the Highly Adaptive LASSO
`Lrnr_hts`	Hierarchical Time-Series Forecasting
`Lrnr_independent_binomial`	Classification from Binomial Regression
`Lrnr_lightgbm`	LightGBM: Light Gradient Boosting Machine
`Lrnr_lstm_keras`	Long short-term memory Recurrent Neural Network (LSTM) with Keras
`Lrnr_mean`	Fitting Intercept Models
`Lrnr_multiple_ts`	Stratify univariable time-series learners by time-series
`Lrnr_multivariate`	Multivariate Learner
`Lrnr_nnet`	Feed-Forward Neural Networks and Multinomial Log-Linear Models
`Lrnr_nnls`	Non-negative Linear Least Squares
`Lrnr_optim`	Optimize Metalearner according to Loss Function using optim
`Lrnr_pca`	Principal Component Analysis and Regression
`Lrnr_polspline`	Polyspline - multivariate adaptive polynomial spline regression (polymars) and polychotomous regression and multiple classification (polyclass)
`Lrnr_pooled_hazards`	Classification from Pooled Hazards
`Lrnr_randomForest`	Random Forests
`Lrnr_ranger`	Ranger: Fast(er) Random Forests
`Lrnr_revere_task`	Learner that chains into a revere task
`Lrnr_rpart`	Learner for Recursive Partitioning and Regression Trees
`Lrnr_rugarch`	Univariate GARCH Models
`Lrnr_screener_augment`	Augmented Covariate Screener
`Lrnr_screener_coefs`	Coefficient Magnitude Screener
`Lrnr_screener_correlation`	Correlation Screening Procedures
`Lrnr_screener_importance`	Variable Importance Screener
`Lrnr_sl`	The Super Learner Algorithm
`Lrnr_solnp`	Nonlinear Optimization via Augmented Lagrange
`Lrnr_solnp_density`	Nonlinear Optimization via Augmented Lagrange
`Lrnr_stratified`	Stratify learner fits by a single variable
`Lrnr_subset_covariates`	Learner with Covariate Subsetting
`Lrnr_svm`	Support Vector Machines
`Lrnr_tsDyn`	Nonlinear Time Series Analysis
`Lrnr_ts_weights`	Time-specific weighting of prediction losses
`Lrnr_xgboost`	xgboost: eXtreme Gradient Boosting
`Pipeline`	Pipeline (chain) of learners.
`Shared_Data`	Container Class for data.table Shared Between Tasks
`Stack`	Learner Stacking
`Lrnr_pkg_SuperLearner`	Use SuperLearner Wrappers, Screeners, and Methods, in sl3
`args_to_list()`	Get all arguments of parent call (both specified and defaults) as list
`bsds`	Bicycle sharing time series dataset
`cpp` `cpp_imputed`	Subset of growth data from the collaborative perinatal project (CPP)
`cpp_1yr`	Subset of growth data from the collaborative perinatal project (CPP)
`train_task()` `validation_task()`	Subset Tasks for CV THe functions use origami folds to subset tasks. These functions are used by Lrnr_cv (and therefore other learners that use Lrnr_cv). So that nested cv works properly, currently the subsetted task objects do not have fold structures of their own, and so generate them from defaults if nested cv is requested.
`cv_risk()`	Cross-validated Risk Estimation
`cv_sl()`	Cross-validated Super Learner
`debug_train()` `debugonce_train()` `debug_predict()` `debugonce_predict()` `sl3_debug_mode()` `undebug_learner()`	Helper functions to debug sl3 Learners
`default_metalearner()`	Automatically Defined Metalearner
`density_dat`	Simulated data with continuous exposure
`factor_to_indicators()` `dt_expand_factors()`	Convert Factors to indicators
`importance()` `importance()`	Importance Extract variable importance measures produced by `randomForest` and order in decreasing order of importance.
`importance_plot()`	Variable Importance Plot
`inverse_sample()`	Inverse CDF Sampling
`delayed_make_learner()` `learner_train()` `delayed_learner_train()` `learner_fit_predict()` `delayed_learner_fit_predict()` `learner_fit_chain()` `delayed_learner_fit_chain()` `learner_subset_covariates()` `learner_process_formula()` `delayed_learner_subset_covariates()` `delayed_learner_process_formula()`	Learner helpers
`sl3_list_properties()` `sl3_list_learners()`	List sl3 Learners
`loss_squared_error()` `loss_loglik_true_cat()` `loss_loglik_binomial()` `loss_loglik_multinomial()` `loss_squared_error_multivariate()`	Loss Function Definitions
`make_learner_stack()`	Make a stack of sl3 learners
`metalearner_logistic_binomial()` `metalearner_linear()` `metalearner_linear_multivariate()` `metalearner_linear_multinomial()`	Combine predictions from multiple learners
`pack_predictions()` `unpack_predictions()`	Pack multidimensional predictions into a vector (and unpack again)
`pooled_hazard_task()`	Generate A Pooled Hazards Task from a Failure Time (or Categorical) Task
`predict_classes()`	Predict Class from Predicted Probabilities
`prediction_plot()`	Plot predicted and true values for diganostic purposes
`process_data()`	Process Data
`risk()`	Risk Estimation
`custom_ROCR_risk()`	FACTORY RISK FUNCTION FOR ROCR PERFORMANCE MEASURES WITH BINARY OUTCOMES
`safe_dim()`	dim that works for vectors too
`sl3Options()`	Querying/setting a single `sl3` option
`make_sl3_Task()`	Define a Machine Learning Task
`sl3_revere_Task`	Revere (SplitSpecific) Task
`subset_folds()`	Make folds work on subset of data
`undocumented_learner`	Undocumented Learner
`variable_type()`	Specify Variable Type
`write_learner_template()`	Generate a file containing a template `sl3` Learner

Reference

Defining A Task

Finding Learners

sl3 Learners

Composing Learners

Loss functions

Metalearner functions

Helpful for Defining Learners

Sample Datasets

Miscellaneous

Index