msgl (version 2.3.9)

cv: Cross Validation

Description

Multinomial sparse group lasso cross validation, with or without parallel backend.

Usage

cv(x, classes, sampleWeights = NULL, grouping = NULL,
  groupWeights = NULL, parameterWeights = NULL, alpha = 0.5,
  standardize = TRUE, lambda, d = 100, fold = 10L,
  cv.indices = list(), intercept = TRUE, sparse.data = is(x,
  "sparseMatrix"), max.threads = NULL, use_parallel = FALSE,
  algorithm.config = msgl.standard.config)

Arguments

x

design matrix, matrix of size \(N \times p\).

classes

classes, factor of length \(N\).

sampleWeights

sample weights, a vector of length \(N\).

grouping

grouping of features (covariates), a vector of length \(p\). Each element of the vector specifying the group of the feature.

groupWeights

the group weights, a vector of length \(m\) (the number of groups). If groupWeights = NULL default weights will be used. Default weights are 0 for the intercept and $$\sqrt{K\cdot\textrm{number of features in the group}}$$ for all other weights.

parameterWeights

a matrix of size \(K \times p\). If parameterWeights = NULL default weights will be used. Default weights are is 0 for the intercept weights and 1 for all other weights.#'

alpha

the \(\alpha\) value 0 for group lasso, 1 for lasso, between 0 and 1 gives a sparse group lasso penalty.

standardize

if TRUE the features are standardize before fitting the model. The model parameters are returned in the original scale.

lambda

lambda.min relative to lambda.max or the lambda sequence for the regularization path.

d

length of lambda sequence (ignored if length(lambda) > 1)

fold

the fold of the cross validation, an integer larger than \(1\) and less than \(N+1\). Ignored if cv.indices != NULL. If fold\(\le\)max(table(classes)) then the data will be split into fold disjoint subsets keeping the ration of classes approximately equal. Otherwise the data will be split into fold disjoint subsets without keeping the ration fixed.

cv.indices

a list of indices of a cross validation splitting. If cv.indices = NULL then a random splitting will be generated using the fold argument.

intercept

should the model include intercept parameters

sparse.data

if TRUE x will be treated as sparse, if x is a sparse matrix it will be treated as sparse by default.

max.threads

Deprecated (will be removed in 2018), instead use use_parallel = TRUE and registre parallel backend (see package 'doParallel'). The maximal number of threads to be used.

use_parallel

If TRUE the foreach loop will use %dopar%. The user must registre the parallel backend.

algorithm.config

the algorithm configuration to be used.

Value

link

the linear predictors -- a list of length length(lambda) one item for each lambda value, with each item a matrix of size \(K \times N\) containing the linear predictors.

response

the estimated probabilities - a list of length length(lambda) one item for each lambda value, with each item a matrix of size \(K \times N\) containing the probabilities.

classes

the estimated classes - a matrix of size \(N \times d\) with \(d=\)length(lambda).

cv.indices

the cross validation splitting used.

features

number of features used in the models.

parameters

number of parameters used in the models.

classes.true

the true classes used for estimation, this is equal to the classes argument

Examples

# NOT RUN {
data(SimData)

# A quick look at the data
dim(x)
table(classes)

# Setup clusters
cl <- makeCluster(2)
registerDoParallel(cl)

# Run cross validation using 2 clusters
# Using a lambda sequence ranging from the maximal lambda to 0.7 * maximal lambda
fit.cv <- msgl::cv(x, classes, alpha = 0.5, lambda = 0.7, use_parallel = TRUE)

# Stop clusters
stopCluster(cl)

# Print some information
fit.cv

# Cross validation errors (estimated expected generalization error)
# Misclassification rate
Err(fit.cv)

# Negative log likelihood error
Err(fit.cv, type="loglike")

# }