lambda: Computes a lambda sequence for the regularization path
In msgl: Multinomial Sparse Group Lasso

Description Usage Arguments Value Author(s) Examples

Computes a decreasing lambda sequence of length d. The sequence ranges from a data determined maximal lambda λ_\textrm{max} to the user inputed lambda.min.

lambda(x, classes, sampleWeights = NULL, grouping = NULL,
  groupWeights = NULL, parameterWeights = NULL, alpha = 0.5,
  d = 100L, standardize = TRUE, lambda.min, intercept = TRUE,
  sparse.data = is(x, "sparseMatrix"), lambda.min.rel = FALSE,
  algorithm.config = msgl.standard.config)

`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, a vector of length p. Each element of the vector specifying the group of the covariate.
`groupWeights`	the group weights, a vector of length m+1 (the number of groups). The first element of the vector is the intercept weight. If `groupWeights = NULL` default weights will be used. Default weights are 0 for the intercept and √{K\cdot\textrm{number of features in the group}} for all other weights.
`parameterWeights`	a matrix of size K \times (p+1). The first column of the matrix is the intercept weights. Default weights are is 0 for the intercept weights and 1 for all other weights.
`alpha`	the α value 0 for group lasso, 1 for lasso, between 0 and 1 gives a sparse group lasso penalty.
`d`	the length of lambda sequence
`standardize`	if TRUE the features are standardize before fitting the model. The model parameters are returned in the original scale.
`lambda.min`	the smallest lambda value in the computed sequence.
`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.
`lambda.min.rel`	is lambda.min relative to lambda.max ? (i.e. actual lambda min used is `lambda.min*lambda.max`, with `lambda.max` the computed maximal lambda value)
`algorithm.config`	the algorithm configuration to be used.

a vector of length d containing the computed lambda sequence.

Martin Vincent

data(SimData)

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

lambda <- msgl::lambda(x, classes, alpha = .5, d = 100, lambda.min = 0.01)

Loading required package: Matrix
Loading required package: sglOptim
Loading required package: foreach
Loading required package: doParallel
Loading required package: iterators
Loading required package: parallel
[1] 100 400
classes
 A  B  C  D  E  F  G  H  I  J 
10 10 10 10 10 10 10 10 10 10