GBM.train {RGBM}R Documentation

Train GBM predictor

Description

This function trains a regression model for a given X.train feature matrix, Y.train response vector, and working parameters. A model returned by this function can be used to predict response for unseen data with GBM.test function.

Usage

GBM.train(X.train, Y.train, s_f = 0.3, s_s = 1, lf =1, M.train = 5000, nu = 0.001)

Arguments

X.train

Input N-by-p feature matrix of training samples. Columns correspond to features, rows correspond to samples.

Y.train

Input N-element response vector of training samples.

s_f

Sampling rate of features, 0<s_f<=1. Fraction of columns from X.train, which will be sampled without replacement to calculate each extesion in boosting model. By default it's 0.3.

s_s

Sampling rate of samples, 0<s_s<=1. Fraction of rows from X.train, which will be sampled with replacement to calculate each extension in boosting model. By default it's 1.

lf

Loss function: 1-> Least Squares and 2 -> Least Absolute Deviation

M.train

Number of extensions in boosting model, e.g. number of iterations of the main loop of RGBM algorithm. By default it's 5000.

nu

Shrinkage factor, learning rate, 0<nu<=1. Each extension to boosting model will be multiplied by the learning rate. By default it's 0.001.

Value

Regression model is a structure containing all the information needed to predict response for unseen data

Author(s)

Raghvendra Mall <rmall@hbku.edu.qa>

See Also

GBM.test


[Package RGBM version 1.0-11 Index]