tuneEnsemble: Train a stacked ensemble of algorithms while tuning its...

In AleMorales/SeedSorter: Supervised classification of seeds from flow cytometry

Description

Train a stacked ensemble of algorithms while tuning its hyperparameters and return the resulting model.

Usage

tuneEnsemble(
  algorithms = c("extinction", "naiveBayes", "lda", "qda"),
  task,
  osw.rate = 10,
  maxiter = 10L,
  lambda = 10L,
  parallel = FALSE,
  nthreads = parallel:::detectCores()
)

Arguments

algorithms	Names of the algorithm to be used as a character vector (same algorithms as in trainAlgorithm()).
task	A classification task as returned by createTrainingTask().
maxiter	Maximum number of iterations in the CMA-ES optimization of hyperparameters.
lambda	Number of offspring in each iteration of the CMA-ES optimization of hyperparameters.
parallel	Whether to use parallelization in the tuning of hyperparameters (default: FALSE).
nthreads	Number of threads/workers to use for parallelization.

Details

The following algorithms can be tuned using CMA-ES optimization: xgboost, logistic, svm and knn.

Value

A trained model that can be used to make predictions.

AleMorales/SeedSorter documentation built on Feb. 12, 2020, 4:13 a.m.