Skip to contents

Default parameters for the SCENIC ExtraTrees regression learner

params_scenic_extra_trees_defaults.Rd

Default parameters for the SCENIC ExtraTrees regression learner

Usage

params_scenic_extra_trees_defaults()

Value

A list with the following parameters:

  • n_trees - Integer. Number of trees to build. Defaults to 500L.

  • min_samples_leaf - Integer. Minimum number of samples required at a leaf node. Defaults to 50L.

  • n_features_split - Integer. Number of features considered per split. 0L resolves to sqrt(n_features) at runtime. Defaults to 0L.

  • n_thresholds - Integer. Number of random thresholds to evaluate per feature per node. Defaults to 1L.

  • max_depth - Integer. Maximum depth of each tree. Defaults to 8L.

  • subsample_frac - Optional numeric. Fraction of cells to subsample per tree. Defaults to NULL.