train_to_final_model: A function that combines cv_loop_train and cv_train_final

In skimlab/CCSBUtils: R Utilitiy Functions for CCSB

A function that combines cv_loop_train and cv_train_final

Description

A function that combines cv_loop_train and cv_train_final

Usage

train_to_final_model(
  data,
  cls,
  stratify = NA,
  train_final = TRUE,
  train_consensus = TRUE,
  fitControl,
  K = 25,
  resampling_rate = 0.8,
  n_features = NA,
  filter_method = c("ROC", "WILCOX"),
  filter_direction = c("two.sided", "greater", "less"),
  filter_threshold_diff = 1,
  filter_threshold_score = 0.8,
  observation_weights = NULL,
  feature_weights = c("uniform", "weighted"),
  predictor_score_threshold = 0.1,
  verbose = FALSE
)

Arguments

data	input matrix, of dimension nobs x nvars; each row is an observation vector. Since this is an input to glmnet, it should be the format that can be used with glmnet
cls	class labels
fitControl	A list of training parameters. See caret::trainControl for detail
K	...
resampling_rate	...
n_features	...
filter_method	...
filter_direction	...
filter_threshold_diff	...
filter_threshold_score	...
observation_weights	...
feature_weights	...
predictor_score_threshold	...

Value

a list of cv_loop_trained (see cv_loop_train_iter), classification_results (see classification_summary_workflow), and final_cv_model (see cv_train_final and additional slots)

skimlab/CCSBUtils documentation built on March 30, 2022, 4:52 a.m.