specmine (version 4.0.0)

train_classifier: Train a classifier

Description

Train a classifier from a dataset object using metadata or data-derived class labels and a resampling strategy supported by caret.

Usage

train_classifier(
  dataset,
  column.class,
  model,
  validation,
  num.folds = 10,
  num.repeats = 10,
  tunelength = 10,
  tunegrid = NULL,
  metric = NULL,
  summary.function = caret::defaultSummary,
  class.in.metadata = TRUE
)

Value

A caret training object returned by caret::train().

Arguments

dataset

A dataset object.

column.class

The metadata column containing the class labels.

model

A model name accepted by caret::train().

validation

Validation method used in training.

num.folds

Number of folds used in resampling.

num.repeats

Number of repeats used in repeated resampling.

tunelength

Number of tuning levels evaluated by caret.

tunegrid

Optional data frame of tuning parameter combinations.

metric

Optional performance metric used for model selection.

summary.function

Summary function passed to caret::trainControl().

class.in.metadata

Logical; if TRUE, class labels are taken from dataset$metadata, otherwise from dataset$data.

Examples

if (FALSE) {
datamat <- matrix(
  rnorm(24),
  nrow = 4,
  dimnames = list(paste0("v", 1:4), paste0("s", 1:6))
)
metadata <- data.frame(class = factor(c("A", "A", "A", "B", "B", "B")))
dataset <- list(data = datamat, metadata = metadata)
train_classifier(dataset, "class", model = "rpart", validation = "cv")
}