train_tsc: Train a time-series classifier

In compstat-lmu/TSClassification: Interface to various time series classifiers from the "The Great Time Series Classification Bake Off"

Description

Set the "java.options" option to use a higher memory if required (e.g. "-Xmx2048m").

Usage

train_tsc(data, target = NULL, classifier, par_vals = NULL,
  model_path = NULL, data_path = NULL, cleanup_data = FALSE)

Arguments

data	character(1) | data.frame Either a path to the dataset or a data.frame that should be saved to disk for modeling. In case a data.frame is provided, the dataset is saved to disk via data_to_path.
target	character(1) Name of the target variable.
classifier	character(1) Character describing the classifier. See ?tsc_classifiers.
par_vals	list (Optional) Hyperparameters for the models. See ?tsc_classifiers for hyperparameters.
model_path	character(1) Path where the resulting model should be saved to.
data_path	character(1) Path where train and test data should be saved to. Defaults to a temporary file.
cleanup_data	logical(1) Should the data be deleted from disk after training / prediction?

Value

NULL, Writes a Java instance of TrainAndPredict to model_path.

Examples

  data = data.frame(matrix(rnorm(300), nrow = 30))
  data$class = factor(sample(letters[1:2], 10, replace = TRUE))
  model_path = tempfile()
  train_tsc(data, target = "class", classifier = "weka.classifiers.trees.J48",
    model_path = model_path)

compstat-lmu/TSClassification documentation built on Nov. 18, 2019, 6:56 a.m.