core.optimization module

Module containing the optimization surrogate.

core.optimization.ref_points

Reference points for benchmark problems.

Type:dict
class core.optimization.Optimization(model)

Bases: object

The class to define the optimization problem.

model

The model object.

Type:core.Model
iterations

Number of optimization iterations.

Type:int
converged

Convergence status.

Type:bool
algorithm

Optization algorithm object.

termination

Termination object.

n_const

NUmber of constraints.

Type:int
ref_point

Reference point for hypervolume calculation.

Type:np.array
direct

Whether a direct optimization is performed.

Type:bool
range_in

Input parameter allowable ranges.

Type:np.array
function

Function used to evaluate the candidates.

problem

Problem object.

Type:datamod.problems.Custom
surrogate

Surrogate object.

Type:core.Surrogate
res

Results object.

Type:pymoo.model.result.Result
optimization_stats

Optimization benchmark statistics.

Type:dict
optimum_model

Candidates evaluated by the original model.

Type:np.array
optimum_surrogate

Candidates evaluated by the surrogate.

Type:np.array
error_measure

Maximum of the error metrics.

Type:float
error

Benchmark percent error.

Type:np.array
benchmark()

Determine the benchmark optimization accuracy.

optimize()

Wrapper function to perform optimization.

plot_results()

Plot the optimized candidates.

report()

Report the optimal solutions and their verification accuracy.

set_problem(surrogate)

Wrapper function to set the problem.

Parameters:surrogate (core.Surrogate) – Surrogate object.

Notes

Direct optimization does not normalize.

verify()

Wrapper function to verify the optimized solutions.