xgboost_boruta: xgboost_boruta

In gm209/gmtools: A collection of generic ML functions for Insurance Risk Modelling

Description

Xgboost flavoured boruta implementation for feature selection. WARN: This method is all brute force so training time can be long.

Usage

xgboost_boruta(df, x, y, w = NULL, niter = 10, xgbParams)

Arguments

df	A data.frame (or coercible object) that contains the training data
x	A list of feature names to be appraised
y	The name of the target column
w	The name of the weights column. defaults to NULL.
niter	How many xgboost models should we train
xgbParams	List of learning parameters to be passed to XGBoost

gm209/gmtools documentation built on May 22, 2019, 2:39 p.m.