Wrapper function to generate DGRDL parameters
params_dgrdl.RdWrapper function to generate DGRDL parameters
Usage
params_dgrdl(
sparsity = 5L,
dict_size = 5L,
alpha = 1,
beta = 1,
max_iter = 20L,
k_neighbours = 5L,
admm_iter = 5L,
rho = 1
)Arguments
- sparsity
Integer. Sparsity constraint (max non-zero coefficients per signal) Defaults to
5L.- dict_size
Integer. Dictionary size Defaults to
5L.- alpha
Numeric. Sample context regularisation weight. Defaults to
1.0.- beta
Numeric. Feature effect regularisation weight. Defaults to
1.0.- max_iter
Integer. Maximum number of iterations for the main algorithm. Defaults to
20L.- k_neighbours
Integer. Number of neighbours in the KNN graph. Defaults to
5L.- admm_iter
Integer. ADMM iterations for sparse coding. Defaults to
5L.- rho
Numeric. ADMM step size. Defaults to
1.0.
Value
A named list with the following elements:
sparsity - Integer. Sparsity constraint (max non-zero coefficients per signal) Defaults to
5L.dict_size - Integer. Dictionary size Defaults to
5L.alpha - Numeric. Sample context regularisation weight. Defaults to
1.0.beta - Numeric. Feature effect regularisation weight. Defaults to
1.0.max_iter - Integer. Maximum number of iterations for the main algorithm. Defaults to
20L.k_neighbours - Integer. Number of neighbours in the KNN graph. Defaults to
5L.admm_iter - Integer. ADMM iterations for sparse coding. Defaults to
5L.rho - Numeric. ADMM step size. Defaults to
1.0.