Control of a ZIPLN fit

Source: R/ZIPLN.R

Helper to define list of parameters to control the PLN fit. All arguments have defaults.

ZIPLN_param(
  backend = c("nlopt"),
  trace = 1,
  covariance = c("full", "diagonal", "spherical", "fixed", "sparse"),
  Omega = NULL,
  penalty = 0,
  penalize_diagonal = TRUE,
  penalty_weights = NULL,
  config_post = list(),
  config_optim = list(),
  inception = NULL
)

Arguments

backend

optimization back used, either "nlopt" or "torch". Default is "nlopt"

trace

a integer for verbosity.

covariance

character setting the model for the covariance matrix. Either "full", "diagonal", "spherical" or "fixed". Default is "full".

Omega

precision matrix of the latent variables. Inverse of Sigma. Must be specified if covariance is "fixed"

penalty

a user-defined penalty to sparsify the residual covariance. Defaults to 0 (no sparsity).

penalize_diagonal

boolean: should the diagonal terms be penalized in the graphical-Lasso? Default is TRUE

penalty_weights

either a single or a list of p x p matrix of weights (default: all weights equal to 1) to adapt the amount of shrinkage to each pairs of node. Must be symmetric with positive values.

config_post

a list for controlling the post-treatments (optional bootstrap, jackknife, R2, etc.). See details

config_optim

a list for controlling the optimizer (either "nlopt" or "torch" backend). See details

inception

Set up the parameters initialization: by default, the model is initialized with a multivariate linear model applied on log-transformed data, and with the same formula as the one provided by the user. However, the user can provide a PLNfit (typically obtained from a previous fit), which sometimes speeds up the inference.

Value

list of parameters used during the fit and post-processing steps

Details

See PLN_param() and PLNnetwork_param() for a full description of the generic optimization parameters. Like PLNnetwork_param(), ZIPLN_param() has two parameters controlling the optimization due the inner-outer loop structure of the optimizer:

  • "ftol_out" outer solver stops when an optimization step changes the objective function by less than ftol_out multiplied by the absolute value of the parameter. Default is 1e-6

  • "maxit_out" outer solver stops when the number of iteration exceeds maxit_out. Default is 100 and one additional parameter controlling the form of the variational approximation of the zero inflation:

  • "approx_ZI" either uses an exact or approximated conditional distribution for the zero inflation. Default is FALSE