Fit POIS-glment — fit_pois_glmnet

The function fits POIS-glment and returns the entire regularization path. No cross-validation is used.

fit_pois_glmnet_nocv(
  z,
  r = max(z),
  nlam = 12,
  lambda = 10^seq(-4, -1.3, length.out = nlam),
  alpha = 1,
  use_parallel = F,
  ncores = 7,
  symmetrize = TRUE
)

Arguments

z	is a potentially sparse count array of dimensions: (sample size) x (data dimension). The entries of z should take values in 0,1,2,...,r.
r	maximum number of levels (K). Defaults to r = max(z).
nlam	number of regularization parameters (lambda) to use; ignored if lambda is provided.
lambda	the vector of regularization parameters to use

Value

A list of fitted POIS models.

Examples

out = fit_pois_glmnet_nocv(amazon, lambda = 10^seq(-4,-1.3, length.out = 5))
#> POIS: glment, global, nocv ... 1.526 (s).