Description Usage Arguments Value
View source: R/gaussian_model_functions.R
Performs the coordinate exchange algorithm for a Scheffé mixture model. It can have many different random starts, or the coordinate exchange algorithm can be performed in a user-supplied matrix.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 |
n_runs |
number of runs |
q |
number of ingredient proportions |
n_random_starts |
number or random starts. Defaults to 100. |
X |
User supplied design matrix. Must be of size (n_runs, q) where: n_runs is the number of runs q is the number of ingredient proportions. |
order |
Order of the Scheffé model (1, 2, or 3). |
opt_method |
Optimization method in each step of the coordinate exchange algorithm. It can be "B" (Brent's algorithm) or "D" (discretization of Cox direction) |
max_it |
integer for maximum number of iterations that the coordinate exchange algorithm will do |
tol |
A positive error tolerance in Brent's method. |
n_cox_points |
number of points to use in the discretization of Cox direction |
plot_designs |
boolean. If TRUE, shows a plot of the initial and the final design. Only works if q is 3. |
verbose |
level of verbosity. |
opt_crit |
optimality criterion: D-optimality ("D" or 0) or I-optimality ("I" or 1) |
seed |
Seed for reproducibility |
n_cores |
Number of cores for parallel processing |
List with the following elements:
X_orig
: The original design. Matrix of size (n_runs, q).
X
: The optimized design. Matrix of size (n_runs, q).
opt_crit_value_orig
: efficiency of the original design.
opt_crit_value
: efficiency of the optimized design.
n_iter
: Number of iterations performed.
efficiency_value_per_iteration
:
opt_crit
: The optimality criterion used.
q
: Number of mixture ingredients.
seed
: seed used to generate the final design. If a design was used as input by the user, this will be NA.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.