sgl_subsampling: Generic sparse group lasso subsampling procedure In sglOptim: Generic Sparse Group Lasso Solver

Description

Subsampling procedure with support parallel computations.

Usage

 1 2 3 4 5 6 sgl_subsampling(module_name, PACKAGE, data, parameterGrouping = NULL, groupWeights = NULL, parameterWeights = NULL, alpha, lambda, d = 100, compute_lambda = length(lambda) == 1, training = NULL, test = NULL, responses = NULL, auto_response_names = TRUE, collapse = FALSE, max.threads = NULL, use_parallel = FALSE, algorithm.config = sgl.standard.config) 

Arguments

 module_name reference to objective specific C++ routines. PACKAGE name of the calling package. data a list of data objects – will be parsed to the specified module. parameterGrouping grouping of parameters, a vector of length p. Each element of the vector specifying the group of the parameters in the corresponding column of β. groupWeights the group weights, a vector of length length(unique(parameterGrouping)) (the number of groups). parameterWeights a matrix of size q \times p. alpha the α value 0 for group lasso, 1 for lasso, between 0 and 1 gives a sparse group lasso penalty. lambda lambda.min relative to lambda.max (if compute_lambda = TRUE) or the lambda sequence for the regularization path, a vector or a list of vectors (of the same length) with the lambda sequence for the subsamples. d length of lambda sequence (ignored if compute_lambda = FALSE) compute_lambda should the lambda sequence be computed training a list of training samples, each item of the list corresponding to a subsample. Each item in the list must be a vector with the indices of the training samples for the corresponding subsample. The length of the list must equal the length of the test list. test a list of test samples, each item of the list corresponding to a subsample. Each item in the list must be vector with the indices of the test samples for the corresponding subsample. The length of the list must equal the length of the training list. responses a vector of responses to simplify and return (if NULL (deafult) no formating will be done) auto_response_names set response names collapse if TRUE the results will be collapsed and ordered into one result, resembling the output of sgl_predict (this is only valid if the test samples are not overlapping) 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.

Details

If no formating is done (i.e. if responses = NULL) then the responses field contains a list of lists structured in the following way:

subsamples 1:

• sample test[[1]][1]

• model (lambda) index 1

• response elements

• model (lambda) index 2

• response elements

• ...

• sample test[[1]][2]

• model (lambda) index 1

• response elements

• model (lambda) index 2

• response elements

• ...

• ...

subsamples 2: ...

If responses = "rname" with rname the name of the response then a list at responses\$rname will be returned. The content of the list will depend on the type of the response.

• vector A list with format subsamples -> models -> matrix of dimension n_i \times q containing the responses for the corresponding model and subsample (where q is the dimension of the response).

• matrix A list with format subsamples -> samples -> models - > the response matrix.

Value

 Y.true the response, that is the y object in data as created by create.sgldata. responses content will depend on the C++ response class features number of features used in the models parameters number of parameters used in the models lambda the lambda sequences used (a vector or list of length length(training)).

Author(s)

Martin Vincent

sglOptim documentation built on Oct. 21, 2018, 9:04 a.m.