CVP_GLASSO: Parallel CV (uses CV_GLASSOc)
In MGallow/GLASSOO: Lasso Penalized Precision Matrix Estimation

Description Usage Arguments Value

Parallel implementation of cross validation.

CVP_GLASSO(X = NULL, lam = 10^seq(-2, 2, 0.2), diagonal = FALSE,
  crit.out = c("avg", "max"), crit.in = c("loss", "avg", "max"),
  tol.out = 1e-04, tol.in = 1e-04, maxit.out = 10000, maxit.in = 10000,
  adjmaxit.out = NULL, K = 5, crit.cv = c("loglik", "AIC", "BIC"),
  start = c("warm", "cold"), cores = 1, trace = c("progress", "print",
  "none"))

`X`	nxp data matrix. Each row corresponds to a single observation and each column contains n observations of a single feature/variable.
`lam`	positive tuning parameters for elastic net penalty. If a vector of parameters is provided, they should be in increasing order. Defaults to grid of values `10^seq(-2, 2, 0.2)`.
`diagonal`	option to penalize the diagonal elements of the estimated precision matrix (Ω). Defaults to `FALSE`.
`crit.out`	criterion for convergence in outer (blockwise) loop. Criterion `avg` will loop until the average absolute parameter change is less than `tol.out` times tolerance multiple. Criterion `max` will loop until the maximum change in the estimated Sigma after an iteration over the parameter set is less than `tol.out`. Defaults to `avg`.
`crit.in`	criterion for convergence in inner (lasso) loop. Criterion for convergence. Criterion `loss` will loop until the relative change in the objective for each response after an iteration is less than `tol.in`. Criterion `avg` will loop until the average absolute change for each response is less than `tol.in` times tolerance multiple. Similary, criterion `max` will loop until the maximum absolute change is less than `tol.in` times tolerance multiple. Defaults to `loss`.
`tol.out`	convergence tolerance for outer (blockwise) loop. Defaults to 1e-4.
`tol.in`	convergence tolerance for inner (lasso) loop. Defaults to 1e-4.
`maxit.out`	maximum number of iterations for outer (blockwise) loop. Defaults to 1e4.
`maxit.in`	maximum number of iterations for inner (lasso) loop. Defaults to 1e4.
`adjmaxit.out`	adjusted maximum number of iterations. During cross validation this option allows the user to adjust the maximum number of iterations after the first `lam` tuning parameter has converged (for each `alpha`). This option is intended to be paired with `warm` starts and allows for 'one-step' estimators. Defaults to NULL.
`K`	specify the number of folds for cross validation.
`crit.cv`	cross validation criterion (`loglik`, `AIC`, or `BIC`). Defaults to `loglik`.
`start`	specify `warm` or `cold` start for cross validation. Default is `warm`.
`cores`	option to run CV in parallel. Defaults to `cores = 1`.
`trace`	option to display progress of CV. Choose one of `progress` to print a progress bar, `print` to print completed tuning parameters, or `none`.

returns list of returns which includes:

`lam`	optimal tuning parameter.
`min.error`	minimum average cross validation error (cv.crit) for optimal parameters.
`avg.error`	average cross validation error (cv.crit) across all folds.
`cv.error`	cross validation errors (cv.crit).