init_xgboost: Initialize a GBM3 brt model

In dahcase/mbgstacking: Stacked generalization ensemble model for model-based geostatistics

Description

Creates an object designed to be passed to init_stacker that describes an gbm3::gbmt_fit model be added to the stacking ensemble.

Creates an object designed to be passed to init_stacker that describes an xgboost::xgb.train model be added to the stacking ensemble.

Usage

init_xgboost(model_name = "xgboost", arguments = list(),
  params_arg = list(nthread = 1), nrounds = 10,
  binomial_evaluation = "prev", weight_column = "")

init_xgboost(model_name = "xgboost", arguments = list(),
  params_arg = list(nthread = 1), nrounds = 10,
  binomial_evaluation = "prev", weight_column = "")

Arguments

model_name	name of the brt model
arguments	named list. Arguments to be passed to the gbm3::gbmt_fit function. Should only be used for options not captured by the params argument in the underlying xgboost function call. See help xgboost::xgb.train for more information
params_arg	named list. Arguments to be passed to the parameters argument of the xgboost::xgb.train function
nrounds	numeric. Max number of iterations
binomial_evaluation	one of 'prev', 'poisson', or 'emplogit'. Prev fits on indicator/N with reg:logistic evaluation. Poisson uses log(N) as an offset while modelling under a poisson. Emplogit transforms things and runs under gaussian family
training_params	named list. Arguments to be passed to the gbm3::training_params function
model_name	name of the brt model
arguments	named list. Arguments to be passed to the xgboost::xgb.train function. Should only be used for options not captured by the params argument in the underlying xgboost function call. See help xgboost::xgb.train for more information
nrounds	numeric. Max number of iterations
binomial_evaluation	one of 'prev', 'poisson', or 'emplogit'. Prev fits on indicator/N with reg:logistic evaluation. Poisson uses log(N) as an offset while modelling under a poisson. Emplogit transforms things and runs under gaussian family

Value

named list of lists with the parameters required to run an brt model

named list of lists with the parameters required to run an brt model

dahcase/mbgstacking documentation built on May 20, 2019, 4:08 p.m.