A Dakota plugin for ropt
The ropt-dakota package extends the ropt
module by providing a plugin that integrates optimization algorithms from the
Dakota toolkit. ropt itself is a robust
optimization framework designed for both continuous and discrete optimization
workflows and is extensible through its plugin architecture. Installing
ropt-dakota makes these Dakota algorithms directly available within ropt.
Reference
Bases: Backend
Dakota optimization backend for ropt.
This class provides an interface to several optimization algorithms from
Dakota, enabling their use within ropt.
This backend cannot run concurrently in-process
Dakota needs a working directory, so a run changes the working directory
of the whole process and writes its input and report files there.
Nothing else may run in that process meanwhile: not a second Dakota run,
not a run on another backend, and not an evaluation function that opens
a file by relative name. To use this backend alongside anything else,
prefix the method with external/ and it runs in a process of its own,
through the external backend.
To select an optimizer, set the method field within the
optimizer section of the
EnOptContext configuration object to the
desired algorithm's name. Most methods support the general options defined
in the EnOptContext object. For
algorithm-specific options, use the options dictionary within the
optimizer section.
Below are the supported options. Click on the common options, or on the
method name to consult the corresponding
Dakota documentation:
Common output option:
Method-specific Options:
| Method | Options |
|---|---|
| optpp_q_newton | search_method, merit_function, steplength_to_boundary, centering_parameter, max_step, gradient_tolerance, max_iterations, convergence_tolerance, max_function_evaluations |
| conmin_mfd | max_iterations, convergence_tolerance, constraint_tolerance, max_function_evaluations |
| conmin_frcg | max_iterations, convergence_tolerance, constraint_tolerance, max_function_evaluations |
| mesh_adaptive_search | initial_delta, variable_tolerance, function_precision, seed, history_file, display_format, variable_neighborhood_search, neighbor_order, display_all_evaluations, use_surrogate, max_iterations, max_function_evaluations |
| coliny_ea | population_size, initialization_type, fitness_type, replacement_type, crossover_rate, crossover_type, mutation_rate, mutation_type, constraint_penalty, solution_target, seed, show_misc_options, misc_options, max_iterations, convergence_tolerance, max_function_evaluations |
| soga | fitness_type, replacement_type, convergence_type, max_iterations, max_function_evaluations, population_size, print_each_pop, initialization_type, crossover_type, crossover_rate, mutation_type, mutation_rate, seed, convergence_tolerance |
| moga | fitness_type, replacement_type, niching_type, convergence_type, postprocessor_type, max_iterations, max_function_evaluations, population_size, print_each_pop, initialization_type, crossover_type, crossover_rate, mutation_type, mutation_rate, seed, convergence_tolerance |
| asynch_pattern_search | initial_delta, contraction_factor, variable_tolerance, solution_target, merit_function, constraint_penalty, smoothing_factor, constraint_tolerance, max_function_evaluations |