Setup hyperparameters for LightRuleFit training.
Usage
setup_LightRuleFit(
nrounds = 200L,
num_leaves = 32L,
max_depth = 4L,
learning_rate = 0.1,
subsample = 0.666,
subsample_freq = 1L,
lambda_l1 = 0,
lambda_l2 = 0,
objective = NULL,
ifw_lightgbm = FALSE,
alpha = 1,
lambda = NULL,
ifw_glmnet = FALSE,
ifw = FALSE
)Arguments
- nrounds
(Tunable) Positive integer: Number of boosting rounds.
- num_leaves
(Tunable) Positive integer: Maximum number of leaves in one tree.
- max_depth
(Tunable) Integer: Maximum depth of trees.
- learning_rate
(Tunable) Numeric: Learning rate.
- subsample
(Tunable) Numeric: Fraction of data to use.
- subsample_freq
(Tunable) Positive integer: Frequency of subsample.
- lambda_l1
(Tunable) Numeric: L1 regularization.
- lambda_l2
(Tunable) Numeric: L2 regularization.
- objective
Character: Objective function.
- ifw_lightgbm
(Tunable) Logical: If TRUE, use Inverse Frequency Weighting in the LightGBM step.
- alpha
(Tunable) Numeric: Alpha for GLMNET.
- lambda
Numeric: Lambda for GLMNET.
- ifw_glmnet
(Tunable) Logical: If TRUE, use Inverse Frequency Weighting in the GLMNET step.
- ifw
Logical: If TRUE, use Inverse Frequency Weighting in classification. This applies IFW to both LightGBM and GLMNET.
Details
Get more information from lightgbm::lgb.train.
Examples
lightrulefit_hyperparams <- setup_LightRuleFit(nrounds = 300L, max_depth = 3L)
lightrulefit_hyperparams
#> <LightRuleFitHyperparameters>
#> hyperparameters:
#> nrounds: <int> 300
#> num_leaves: <int> 32
#> max_depth: <int> 3
#> learning_rate: <nmr> 0.10
#> subsample: <nmr> 0.67
#> subsample_freq: <int> 1
#> lambda_l1: <nmr> 0.00
#> lambda_l2: <nmr> 0.00
#> objective: <NUL> NULL
#> ifw_lightgbm: <lgc> FALSE
#> alpha: <nmr> 1.00
#> lambda: <NUL> NULL
#> ifw_glmnet: <lgc> FALSE
#> ifw: <lgc> FALSE
#> tunable_hyperparameters: <chr> nrounds, num_leaves, max_depth, learning_rate, subsample, subsample_freq, lambda_l1, lambda_l2, alpha, ifw_lightgbm, ifw_glmnet
#> fixed_hyperparameters: <chr> lambda, objective
#> tuned: <int> -1
#> resampled: <int> 0
#> n_workers: <int> 1
#>
#> No search values defined for tunable hyperparameters.