Skip to contents

Setup ClusterConfig

Source: R/130_ClusterConfig.R
setup_ClusterConfig.Rd

Setup a ClusterConfig object. ClusterConfig is a portable, data-agnostic recipe for a clustering pipeline: dat_path is optional, so the same config can be validated, shared, or described without data and have a path bound later (e.g. by the rtemis CLI) before cluster.

Usage

setup_ClusterConfig(
  dat_path = NULL,
  clustering_config = NULL,
  execution_config = setup_FutureExecution(),
  outdir = "results/",
  verbosity = 1L
)

Arguments

dat_path

Character or NULL: Path to input data file. NULL leaves the recipe unbound; set it (or supply data) before cluster.

clustering_config

ClusteringConfig object or NULL: Configuration for the clustering itself, naming the algorithm, its settings and the feature columns to cluster on. Setup with a clustering setup_* function, e.g. setup_KMeans. If NULL, cluster runs its default algorithm with default settings.

execution_config

ExecutionConfig object: Execution settings, e.g. setup_FutureExecution. Its seed seeds the fit.

outdir

Character: Output directory for results.

verbosity

Integer [0, Inf): Verbosity level.

Value

ClusterConfig object.

Author

EDG

Examples

cc <- setup_ClusterConfig(
  dat_path = "data.csv",
  clustering_config = setup_KMeans(k = 3L),
  outdir = "results/"
)