Package index • nadir

Fitting super learners

super_learner(): Super Learner: Cross-Validation Based Ensemble Learning
cv_super_learner(): Cross-Validating a super_learner
crossfit_super_learner(): Cross-Fit a Super Learner and Retain Fold-Specific Predictors
compare_learners(): Compare Learners

Learners

lnr_bart(): Bayesian Additive Regression Trees (BART) Learner
lnr_cvglmnet(): cv.glmnet Learner
lnr_earth(): Earth Learner
lnr_gam(): Generalized Additive Model Learner
lnr_gausspr(): Gaussian Process Regression Learner
lnr_gbm(): Gradient Boosting Machines Learner
lnr_glm(): GLM Learner
lnr_glm_density(): Conditional Normal Density Estimation Given Mean Predictors — with GLMs
lnr_glmer(): Generalized Linear Mixed-Effects (lme4::glmer) Learner
lnr_glmnet(): glmnet Learner
lnr_glmnet_grid(): glmnet Learner over a Grid of Lambda Values
lnr_hal(): Highly Adaptive Lasso
lnr_hal_grid(): Highly Adaptive Lasso over a Grid of Lambda Values
lnr_heteroskedastic_density(): Conditional Density Estimation with Heteroskedasticity
lnr_homoskedastic_density(): Conditional Density Estimation with Homoskedasticity Assumption
lnr_knn(): k-Nearest Neighbors Learner
lnr_knn_binary(): k-Nearest Neighbors Learner for Binary Classification
lnr_lightgbm(): LightGBM Learner
lnr_lm(): Linear Model Learner
lnr_lm_density(): Conditional Normal Density Estimation Given Mean Predictors
lnr_lmer(): Random/Mixed-Effects (lme4::lmer) Learner
lnr_logistic(): Standard Logistic Regression for Binary Classification
lnr_mean(): Mean Learner
lnr_multinomial_nnet(): nnet::multinom Multinomial Learner
lnr_multinomial_ranger(): ranger Multinomial Learner
lnr_multinomial_vglm(): VGAM::vglm Multinomial Learner
lnr_nnet(): Use nnet for Binary Classification
lnr_ppr(): Projection Pursuit Regression Learner
lnr_ranger(): ranger Learner
lnr_ranger_binary(): ranger Learner for Binary Outcomes
lnr_rf(): randomForest Learner
lnr_rf_binary(): Use Random Forest for Binary Classification
lnr_rpart(): Recursive Partitioning (CART) Learner
lnr_rpart_binary(): Recursive Partitioning (CART) Learner for Binary Classification
lnr_svm(): Support Vector Machine Learner
lnr_svm_binary(): Support Vector Machine Learner for Binary Classification
lnr_xgboost(): XGBoost Learner
learners: Learners in the nadir Package
binary_learners: Binary Learners in nadir
density_learners: Conditional Density Estimation in the nadir Package
multiclass_learners: Multiclass Learners in nadir
list_known_learners(): List Known Learners
add_screener(): Add a Screener to a Learner
truncate_lnr(): Helper to Truncate a Learner's Predictions

Cross-validation schemas

cv_character_and_factors_schema(): Cross Validation Training/Validation Splits with Characters/Factor Columns
cv_origami_schema(): Cross-Validation with Origami
cv_random_schema(): Assign Data to One of n_folds Randomly and Produce Training/Validation Data Lists
cv_super_learner(): Cross-Validating a super_learner
add_stratification(): Stratify a Learner by One or More Categorical Variables

Model methods

print(<nadir_sl_model>) summary(<nadir_sl_model>) coef(<nadir_sl_model>) fitted(<nadir_sl_model>) residuals(<nadir_sl_model>) formula(<nadir_sl_model>) nobs(<nadir_sl_model>) predict(<nadir_sl_model>): Print a nadir_sl_model
predict(<nadir_crossfit_sl>) coef(<nadir_crossfit_sl>) fitted(<nadir_crossfit_sl>) residuals(<nadir_crossfit_sl>) formula(<nadir_crossfit_sl>) nobs(<nadir_crossfit_sl>) summary(<nadir_crossfit_sl>) print(<summary.nadir_crossfit_sl>) plot(<nadir_crossfit_sl>): Predicting from a Cross-Fitted Super Learner
plot(<nadir_cv_sl>) print(<nadir_cv_sl>): Plot a Cross-Validated Super Learner
plot(<nadir_sl_model>): Plot a nadir_sl_model

Weights, losses, and screeners

determine_weights_for_binary_outcomes(): Determine Weights Appropriately for Super Learner given Binary Outcomes
determine_weights_using_neg_log_loss(): Determine Weights for Density Estimators for SuperLearner
determine_super_learner_weights_nnls(): Determine SuperLearner Weights with Nonnegative Least Squares
negative_log_loss(): Negative Log Loss
negative_log_loss_for_binary(): Negative Log Loss for Binary
mse(): Mean Squared Error
screener_cor(): Correlation Threshold Based Screening
screener_cor_top_n(): Correlation Threshold Based Screening
screener_t_test(): t-test Based Screening: Thresholds on p.values and/or t statistics
screeners: Wrapping Learners with a Screener

Utilities

nadir_supported_types: Outcome types supported by nadir
as_multi_predictor(): Construct a Multi-Predictor
df_to_survival_stacked(): Repeat Observations for Survival Stacking
make_learner_names_unique(): Make Unique Learner Names