glmnetUtils (version 1.1.3)

cv.glmnet: Formula interface for elastic net cross-validation with cv.glmnet

Description

Formula interface for elastic net cross-validation with cv.glmnet

Usage

cv.glmnet(x, ...)

# S3 method for default cv.glmnet(x, y, ...)

# S3 method for formula cv.glmnet(formula, data, alpha = 1, nfolds = 10, ..., weights = NULL, offset = NULL, subset = NULL, na.action = getOption("na.action"), drop.unused.levels = FALSE, xlev = NULL, sparse = FALSE, use.model.frame = FALSE, gamma = c(0, 0.25, 0.5, 0.75, 1), relax = FALSE)

# S3 method for cv.glmnet.formula predict(object, newdata, na.action = na.pass, ...)

# S3 method for cv.glmnet.formula coef(object, ...)

# S3 method for cv.glmnet.formula print(x, ...)

# S3 method for cv.relaxed.formula predict(object, newdata, na.action = na.pass, ...)

# S3 method for cv.glmnet.formula coef(object, ...)

Arguments

x

For the default method, a matrix of predictor variables.

...

For cv.glmnet.formula and cv.glmnet.default, other arguments to be passed to glmnet::cv.glmnet; for the predict and coef methods, arguments to be passed to their counterparts in package glmnet.

y

For the default method, a response vector or matrix (for a multinomial response).

formula

A model formula; interaction terms are allowed and will be expanded per the usual rules for linear models.

data

A data frame or matrix containing the variables in the formula.

alpha

The elastic net mixing parameter. See glmnet::glmnet for more details.

nfolds

The number of crossvalidation folds to use. See glmnet::cv.glmnet for more details.

weights

An optional vector of case weights to be used in the fitting process. If missing, defaults to an unweighted fit.

offset

An optional vector of offsets, an a priori known component to be included in the linear predictor.

subset

An optional vector specifying the subset of observations to be used to fit the model.

na.action

A function which indicates what should happen when the data contains missing values. For the predict method, na.action = na.pass will predict missing values with NA; na.omit or na.exclude will drop them.

drop.unused.levels

Should factors have unused levels dropped? Defaults to FALSE.

xlev

A named list of character vectors giving the full set of levels to be assumed for each factor.

sparse

Should the model matrix be in sparse format? This can save memory when dealing with many factor variables, each with many levels.

use.model.frame

Should the base model.frame function be used when constructing the model matrix? This is the standard method that most R modelling functions use, but has some disadvantages. The default is to avoid model.frame and construct the model matrix term-by-term; see discussion.

gamma

For cv.glmnet.formula, the values of the parameter for mixing the relaxed (non-regularised) fit with the regularized fit. Not used if relax=FALSE. Requires glmnet 3.0 or later.

relax

For cv.glmnet.formula, whether to perform a relaxed fit after the regularised one. Requires glmnet 3.0 or later.

object

For the predict and coef methods, an object of class cv.glmnet.formula.

newdata

For the predict method, a data frame containing the observations for which to calculate predictions.

Value

For cv.glmnet.formula, an object of class either cv.glmnet.formula or cv.relaxed.formula, based on the value of the relax argument. This is basically the same object created by glmnet::cv.glmnet, but with extra components to allow formula usage.

Details

The cv.glmnet function in this package is an S3 generic with a formula and a default method. The former calls the latter, and the latter is simply a direct call to the cv.glmnet function in package glmnet. All the arguments to glmnet::cv.glmnet are (or should be) supported.

There are two ways in which the matrix of predictors can be generated. The default, with use.model.frame = FALSE, is to process the additive terms in the formula independently. With wide datasets, this is much faster and more memory-efficient than the standard R approach which uses the model.frame and model.matrix functions. However, the resulting model object is not exactly the same as if the standard approach had been used; in particular, it lacks a bona fide terms object. If you require interoperability with other packages that assume the standard model object structure, set use.model.frame = TRUE. See discussion for more information on this topic.

The predict and coef methods are wrappers for the corresponding methods in the glmnet package. The former constructs a predictor model matrix from its newdata argument and passes that as the newx argument to glmnet:::predict.cv.glmnet.

See Also

glmnet::cv.glmnet, glmnet::predict.cv.glmnet, glmnet::coef.cv.glmnet, model.frame, model.matrix

Examples

Run this code
# NOT RUN {
cv.glmnet(mpg ~ ., data=mtcars)

cv.glmnet(Species ~ ., data=iris, family="multinomial")

# }
# NOT RUN {
# Leukemia example dataset from Trevor Hastie's website
download.file("https://web.stanford.edu/~hastie/glmnet/glmnetData/Leukemia.RData",
              "Leukemia.RData")
load("Leukemia.Rdata")
leuk <- do.call(data.frame, Leukemia)
cv.glmnet(y ~ ., leuk, family="binomial")
# }

Run the code above in your browser using DataCamp Workspace