predict.hal: predict.hal

In benkeser/halplus: Highly Adaptive Lasso

Description

A predict method for objects of class hal.

Usage

## S3 method for class 'hal'
predict(object, newdata, bigDesign = FALSE, verbose = TRUE,
  chunks = 5000, s = ifelse(object$useMin, object$object$lambda.min,
  object$object$lambda.1se), ...)

Arguments

object	A fitted object of class hal
newdata	A matrix of new predictions to obtain predictions
bigDesign	A boolean indicating whether to obtain predictions all at once (which may be memory intractable) or to split up the task into smaller chunks
verbose	A boolean indicating whether to print output on functions progress
chunks	A numeric indicating how many chunks to split the prediction task into (if bigDesign = FALSE)
s	Option from glmnet indicating what value to use. Defaults to the value that was specified in the original call to hal (which itself defaults to the value which minimizes MSE).
...	Additional arguments (not currently used)

benkeser/halplus documentation built on May 12, 2019, noon