predict.relaxnet: Predict Method for '"relaxnet"' Objects

In relaxnet: Relaxation of glmnet models (as in relaxed lasso, Meinshausen 2007)

Description

Similar to other predict methods, this functions predicts fitted values, logits, coefficients and more from a fitted "relaxnet" object.

Usage

## S3 method for class 'relaxnet'
predict(object,
        newx,
        which.model,
        s = NULL,
        type = c("link", "response", "coefficients", "nonzero", "class"),
        exact = FALSE,
	...)

Arguments

object	The "relaxnet" object from which to get predictions.
newx	Matrix of new values for x at which predictions are to be made. Must be a matrix; can be sparse as in Matrix package. This argument is not used for type=c("coefficients","nonzero")
which.model	Specifies the submodel from which predictions are required. "main" indicates the main glmnet model, while an integer indicates one of the relaxed models.
s	Value(s) of the penalty parameter lambda at which predictions are required. Default is the entire sequence of lambda values for the model specified by which.model.
type	See link[glmnet]{predict.glmnet}.
exact	Only the default, FALSE, is supported. See link[glmnet]{predict.glmnet}.
...	Further arguments passed to predict.glmnet. In the current version, these are not guaranteed to work correctly (for example, offset has not yet been implemented for relaxnet).

Value

The object returned depends on type.

Author(s)

Stephan Ritter, with design contributions from Alan Hubbard.

Much of the code (and some help file content) is adapted from the glmnet package, whose authors are Jerome Friedman, Trevor Hastie and Rob Tibshirani.

See Also

relaxnet, predict.glmnet

relaxnet documentation built on May 2, 2019, 12:39 p.m.