doe_lib module¶
DOE library base class wrapper¶
Classes:
Abstract class to use for DOE library link See DriverLib. |
- class gemseo.algos.doe.doe_lib.DOELibrary[source]¶
Bases:
gemseo.algos.driver_lib.DriverLibAbstract class to use for DOE library link See DriverLib.
Constructor Abstract class.
Attributes:
Return the available algorithms.
Methods:
compute_phip_criteria(samples)Compute the phi^p criteria (see Morris & Mitchell, Exploratory designs for computational experiments, 1995)
driver_has_option(option_key)Checks if the option key exists.
ensure_bounds(orig_func[, normalize])Project the design vector onto the design space before execution.
evaluate_samples([eval_jac, n_processes, …])Evaluates all functions of optimization problem at the samples.
execute(problem[, algo_name])Executes the driver.
export_samples(doe_output_file)Export samples generated by DOE library to a csv file.
filter_adapted_algorithms(problem)Filters the algorithms capable of solving the problem.
Finalize the iteration observer.
get_optimum_from_database([message, status])Retrieves the optimum from the database and builds an optimization result object from it.
get_x0_and_bounds_vects(normalize_ds)Gets x0, bounds, normalized or not depending on algo options, all as numpy arrays.
init_iter_observer(max_iter, message)Initialize the iteration observer.
init_options_grammar(algo_name)Initializes the options grammar.
is_algo_requires_grad(algo_name)Returns True if the algorithm requires a gradient evaluation.
is_algorithm_suited(algo_dict, problem)Checks if the algorithm is suited to the problem according to its algo dict.
Callback called at each new iteration, ie every time a design vector that is not already in the database is proposed by the optimizer.
- COMPLEX_STEP_METHOD = 'complex_step'¶
- DESCRIPTION = 'description'¶
- DESIGN_ALGO_NAME = 'Design algorithm'¶
- DIFFERENTIATION_METHODS = ['user', 'complex_step', 'finite_differences']¶
- DIMENSION = 'dimension'¶
- EQ_TOLERANCE = 'eq_tolerance'¶
- EVAL_JAC = 'eval_jac'¶
- FINITE_DIFF_METHOD = 'finite_differences'¶
- HANDLE_EQ_CONS = 'handle_equality_constraints'¶
- HANDLE_INEQ_CONS = 'handle_inequality_constraints'¶
- INEQ_TOLERANCE = 'ineq_tolerance'¶
- INTERNAL_NAME = 'internal_algo_name'¶
- LIB = 'lib'¶
- MAX_DS_SIZE_PRINT = 40¶
- MAX_TIME = 'max_time'¶
- MIN_DIMS = 'min_dims'¶
- NORMALIZE_DESIGN_SPACE_OPTION = 'normalize_design_space'¶
- N_PROCESSES = 'n_processes'¶
- N_SAMPLES = 'n_samples'¶
- OPTIONS_DIR = 'options'¶
- OPTIONS_MAP = {}¶
- PHIP_CRITERIA = 'phi^p'¶
- POSITIVE_CONSTRAINTS = 'positive_constraints'¶
- PROBLEM_TYPE = 'problem_type'¶
- REQUIRE_GRAD = 'require_grad'¶
- ROUND_INTS_OPTION = 'round_ints'¶
- SAMPLES_TAG = 'samples'¶
- SEED = 'seed'¶
- USER_DEFINED_GRADIENT = 'user'¶
- USE_DATABASE_OPTION = 'use_database'¶
- WAIT_TIME_BETWEEN_SAMPLES = 'wait_time_between_samples'¶
- WEBSITE = 'website'¶
- property algorithms¶
Return the available algorithms.
- static compute_phip_criteria(samples)[source]¶
Compute the phi^p criteria (see Morris & Mitchell, Exploratory designs for computational experiments, 1995)
- Parameters
samples – design variables list
- driver_has_option(option_key)¶
Checks if the option key exists.
- Parameters
option_key – the name of the option
- Returns
True if the option is in the grammar
- ensure_bounds(orig_func, normalize=True)¶
Project the design vector onto the design space before execution.
- Parameters
orig_func – the original function
normalize – if True, use the normalized design space
- Returns
the wrapped function
- evaluate_samples(eval_jac=False, n_processes=1, wait_time_between_samples=0)[source]¶
Evaluates all functions of optimization problem at the samples.
- Parameters
eval_jac – if True, the jacobian is also evaluated (Default value = False)
- execute(problem, algo_name=None, **options)¶
Executes the driver.
- Parameters
problem – the problem to be solved
algo_name – name of the algorithm if None, use self.algo_name which may have been set by the factory (Default value = None)
options – the options dict for the algorithm
- export_samples(doe_output_file)[source]¶
Export samples generated by DOE library to a csv file.
- Parameters
doe_output_file (string) – export file name
- filter_adapted_algorithms(problem)¶
Filters the algorithms capable of solving the problem.
- Parameters
problem – the opt_problem to be solved
- Returns
the list of adapted algorithms names
- finalize_iter_observer()¶
Finalize the iteration observer.
- get_optimum_from_database(message=None, status=None)¶
Retrieves the optimum from the database and builds an optimization result object from it.
- Parameters
message – Default value = None)
status – Default value = None)
- get_x0_and_bounds_vects(normalize_ds)¶
Gets x0, bounds, normalized or not depending on algo options, all as numpy arrays.
- Parameters
normalize_ds – if True, normalizes all input vars that are not integers, according to design space normalization policy
- Returns
x, lower bounds, upper bounds
- init_iter_observer(max_iter, message)¶
Initialize the iteration observer.
It will handle the stopping criterion and the logging of the progress bar.
- Parameters
max_iter – maximum number of calls
message – message to display at the beginning
- init_options_grammar(algo_name)¶
Initializes the options grammar.
- Parameters
algo_name – name of the algorithm
- is_algo_requires_grad(algo_name)¶
Returns True if the algorithm requires a gradient evaluation.
- Parameters
algo_name – name of the algorithm
- static is_algorithm_suited(algo_dict, problem)[source]¶
Checks if the algorithm is suited to the problem according to its algo dict.
- Parameters
algo_dict – the algorithm characteristics
problem – the opt_problem to be solved
- new_iteration_callback()¶
Callback called at each new iteration, ie every time a design vector that is not already in the database is proposed by the optimizer.
Iterates the progress bar, implements the stop criteria.