setup.LightRuleFit: Set s_LightRuleFit parameters
In egenn/rtemis: Machine Learning and Visualization

setup.LightRuleFit

R Documentation

Set s_LightRuleFit parameters

Description

Sets parameters for the GBM and GLMNET (LASSO) steps of s_LightRuleFit

Usage

setup.LightRuleFit(
  n_trees = 200,
  num_leaves = 32L,
  max_depth = 3,
  learning_rate = 0.1,
  subsample = 0.666,
  subsample_freq = 1L,
  lambda_l1 = 0,
  lambda_l2 = 0,
  objective = NULL,
  extra.lgbm.params = NULL,
  lightgbm.ifw = TRUE,
  lightgbm.resample.params = setup.resample("kfold", 5),
  glmnet.ifw = TRUE,
  importance = FALSE,
  alpha = 1,
  lambda = NULL,
  glmnet.resample.params = setup.resample("kfold", 5)
)

Arguments

`num_leaves`	Integer: [gS] Maximum tree leaves for base learners.
`max_depth`	Integer: [gS] Maximum tree depth for base learners, <=0 means no limit.
`learning_rate`	Numeric: [gS] Boosting learning rate
`subsample`	Numeric: [gS] Subsample ratio of the training set.
`subsample_freq`	Integer: Subsample every this many iterations
`lambda_l1`	Numeric: [gS] L1 regularization term
`lambda_l2`	Numeric: [gS] L2 regularization term
`objective`	(Default = NULL)
`lightgbm.ifw`	Logical: Passed to s_LightGBM's `ifw` argument
`glmnet.ifw`	Logical: Passed to s_GLMNET's `ifw` argument
`importance`	Logical: If `TRUE`, calculate variable importance
`alpha`	[gS] Float [0, 1]: The elasticnet mixing parameter: `a = 0` is the ridge penalty, `a = 1` is the lasso penalty
`lambda`	[gS] Float vector: Best left to NULL, `cv.glmnet` will compute its own lambda sequence