predict.gbts {gbts}R Documentation

Predict method for ensemble of Gradient Boosted Trees

Description

This function generates predictions by weighted averaging the predictions from each model in the ensemble returned from gbts. Weighted average is computed on the log-odds scale for binary classification.

Usage

## S3 method for class 'gbts'
predict(object, x, nwrk = 2, ...)

Arguments

object

a model object returned from gbts.

x

a data.frame of predictors. It must follow the same format as the training dataset on which the model object is developed.

nwrk

an integer of the number of computing workers to be used. If nwrk is less than the number of available cores on the machine, it uses all available cores.

...

further arguments passed to or from other methods.

Value

A numeric vector of predictions. In the case of binary classification, predictions are probabilities.

Author(s)

Waley W. J. Liang <wliang10@gmail.com>

See Also

gbts, comperf


[Package gbts version 1.2.0 Index]