pyleecan.Classes.OptiBayesAlgSmoot module

Method code available at https://github.com/Eomys/pyleecan/tree/master/pyleecan/Methods/Optimization/OptiBayesAlgSmoot

class OptiBayesAlgSmoot(size_pop=40, nb_gen=100, nb_iter=15, nb_start=300, criterion='PI', kernel=0, problem=- 1, xoutput=- 1, logger_name='Pyleecan.OptiSolver', is_keep_all_output=False, init_dict=None, init_str=None)[source]

Bases: OptiBayesAlg

Multi-objectives optimization problem with some constraints

VERSION = 1
solve()

Method to perform Bayesian optimization using Smoot tools

Parameters:

self (OptiBayesAlgSmoot) – Solver to perform Bayesian model creation, then use a genetic algorithm

Returns:

multi_output – class containing the results

Return type:

OutputMultiOpti

check_optimization_input()

Check optimization parameters before solving the problem

Parameters:

solver (Solver) – solver to perform the bayesian algorithm with SMT

evaluate(input_x)
plot_pareto(x_symbol, y_symbol, c_symbol=None, cmap=None, ax=None, title=None, grid=False, is_show_fig=True, save_path=None)

Plot the pareto front for 2 objective functions

Parameters:

  • self (OptiBayesAlgSmoot) –

  • x_symbol (str) – symbol of the first objective function

  • y_symbol (str) – symbol of the second objective function

  • c_symbol (str) – optional symbol to set the plot colors

  • cmap (colormap) – optional colormap

  • is_show_fig (bool) – True to show figure after plot

  • save_path (str) – full path of the png file where the figure is saved if save_path is not None

eval_const(constraint, input_x)

Evaluate the constraint at given points x

save(save_path='', is_folder=False, type_handle_old=2, type_compression=0)

Save the object to the save_path

Parameters:

  • self – A pyleecan object

  • save_path (str) – path to the folder to save the object

  • is_folder (bool) – to split the object in different files: separate simulation machine and materials (json only)

  • type_handle_old (int) – How to handle old file in folder mode (0:Nothing, 1:Delete, 2:Move to “Backup” folder)

  • type_compression (int) – Available only for json, 0: no compression, 1: gzip

get_logger()

Get the object logger or its parent’s one

Parameters:

obj – A pyleecan object

Returns:

logger – Pyleecan object dedicated logger

Return type:

logging.Logger

compare(other, name='self', ignore_list=None, is_add_value=False)[source]

Compare two objects and return list of differences

as_dict(type_handle_ndarray=0, keep_function=False, **kwargs)[source]

Convert this object in a json serializable dict (can be use in __init__). type_handle_ndarray: int

How to handle ndarray (0: tolist, 1: copy, 2: nothing)

keep_functionbool

True to keep the function object, else return str

Optional keyword input parameter is for internal use only and may prevent json serializability.

copy()[source]

Creates a deepcopy of the object

property size_pop

Number of individuals for each generation

Type:

int

Min:

1

property nb_gen

Number of generations for the genetic part

Type:

int

Min:

1